diff options
Diffstat (limited to 'kernel')
80 files changed, 1715 insertions, 1206 deletions
diff --git a/kernel/Makefile b/kernel/Makefile index aa7368c7eabf..320f1f3941b7 100644 --- a/kernel/Makefile +++ b/kernel/Makefile @@ -51,7 +51,7 @@ obj-y += livepatch/ obj-y += dma/ obj-y += entry/ -obj-$(CONFIG_CHECKPOINT_RESTORE) += kcmp.o +obj-$(CONFIG_KCMP) += kcmp.o obj-$(CONFIG_FREEZER) += freezer.o obj-$(CONFIG_PROFILING) += profile.o obj-$(CONFIG_STACKTRACE) += stacktrace.o diff --git a/kernel/audit.c b/kernel/audit.c index 1ffc2e059027..551a394bc8f4 100644 --- a/kernel/audit.c +++ b/kernel/audit.c @@ -2285,7 +2285,7 @@ static void audit_log_set_loginuid(kuid_t koldloginuid, kuid_t kloginuid, uid = from_kuid(&init_user_ns, task_uid(current)); oldloginuid = from_kuid(&init_user_ns, koldloginuid); - loginuid = from_kuid(&init_user_ns, kloginuid), + loginuid = from_kuid(&init_user_ns, kloginuid); tty = audit_get_tty(); audit_log_format(ab, "pid=%d uid=%u", task_tgid_nr(current), uid); @@ -2365,7 +2365,7 @@ int audit_signal_info(int sig, struct task_struct *t) * * We can not do a netlink send inside an irq context because it blocks (last * arg, flags, is not set to MSG_DONTWAIT), so the audit buffer is placed on a - * queue and a tasklet is scheduled to remove them from the queue outside the + * queue and a kthread is scheduled to remove them from the queue outside the * irq context. May be called in any context. */ void audit_log_end(struct audit_buffer *ab) diff --git a/kernel/audit_fsnotify.c b/kernel/audit_fsnotify.c index 5b3f01da172b..60739d5e3373 100644 --- a/kernel/audit_fsnotify.c +++ b/kernel/audit_fsnotify.c @@ -84,7 +84,7 @@ struct audit_fsnotify_mark *audit_alloc_mark(struct audit_krule *krule, char *pa dentry = kern_path_locked(pathname, &path); if (IS_ERR(dentry)) - return (void *)dentry; /* returning an error */ + return ERR_CAST(dentry); /* returning an error */ inode = path.dentry->d_inode; inode_unlock(inode); diff --git a/kernel/auditsc.c b/kernel/auditsc.c index ce8c9e2279ba..47fb48f42c93 100644 --- a/kernel/auditsc.c +++ b/kernel/auditsc.c @@ -799,12 +799,12 @@ static int audit_in_mask(const struct audit_krule *rule, unsigned long val) return rule->mask[word] & bit; } -/* At syscall entry and exit time, this filter is called if the - * audit_state is not low enough that auditing cannot take place, but is - * also not high enough that we already know we have to write an audit - * record (i.e., the state is AUDIT_SETUP_CONTEXT or AUDIT_BUILD_CONTEXT). +/* At syscall exit time, this filter is called if the audit_state is + * not low enough that auditing cannot take place, but is also not + * high enough that we already know we have to write an audit record + * (i.e., the state is AUDIT_SETUP_CONTEXT or AUDIT_BUILD_CONTEXT). */ -static enum audit_state audit_filter_syscall(struct task_struct *tsk, +static void audit_filter_syscall(struct task_struct *tsk, struct audit_context *ctx, struct list_head *list) { @@ -812,7 +812,7 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk, enum audit_state state; if (auditd_test_task(tsk)) - return AUDIT_DISABLED; + return; rcu_read_lock(); list_for_each_entry_rcu(e, list, list) { @@ -821,11 +821,11 @@ static enum audit_state audit_filter_syscall(struct task_struct *tsk, &state, false)) { rcu_read_unlock(); ctx->current_state = state; - return state; + return; } } rcu_read_unlock(); - return AUDIT_BUILD_CONTEXT; + return; } /* @@ -1930,7 +1930,7 @@ static inline int audit_copy_fcaps(struct audit_names *name, if (!dentry) return 0; - rc = get_vfs_caps_from_disk(dentry, &caps); + rc = get_vfs_caps_from_disk(&init_user_ns, dentry, &caps); if (rc) return rc; @@ -2481,7 +2481,8 @@ int __audit_log_bprm_fcaps(struct linux_binprm *bprm, ax->d.next = context->aux; context->aux = (void *)ax; - get_vfs_caps_from_disk(bprm->file->f_path.dentry, &vcaps); + get_vfs_caps_from_disk(&init_user_ns, + bprm->file->f_path.dentry, &vcaps); ax->fcap.permitted = vcaps.permitted; ax->fcap.inheritable = vcaps.inheritable; diff --git a/kernel/bpf/btf.c b/kernel/bpf/btf.c index c914d68f3bf8..369faeddf1df 100644 --- a/kernel/bpf/btf.c +++ b/kernel/bpf/btf.c @@ -4400,8 +4400,6 @@ btf_get_prog_ctx_type(struct bpf_verifier_log *log, struct btf *btf, * is not supported yet. * BPF_PROG_TYPE_RAW_TRACEPOINT is fine. */ - if (log->level & BPF_LOG_LEVEL) - bpf_log(log, "arg#%d type is not a struct\n", arg); return NULL; } tname = btf_name_by_offset(btf, t->name_off); diff --git a/kernel/bpf/core.c b/kernel/bpf/core.c index 0ae015ad1e05..3a283bf97f2f 100644 --- a/kernel/bpf/core.c +++ b/kernel/bpf/core.c @@ -1118,6 +1118,8 @@ static void bpf_prog_clone_free(struct bpf_prog *fp) * clone is guaranteed to not be locked. */ fp->aux = NULL; + fp->stats = NULL; + fp->active = NULL; __bpf_prog_free(fp); } @@ -2342,6 +2344,10 @@ bool __weak bpf_helper_changes_pkt_data(void *func) /* Return TRUE if the JIT backend wants verifier to enable sub-register usage * analysis code and wants explicit zero extension inserted by verifier. * Otherwise, return FALSE. + * + * The verifier inserts an explicit zero extension after BPF_CMPXCHGs even if + * you don't override this. JITs that don't want these extra insns can detect + * them using insn_is_zext. */ bool __weak bpf_jit_needs_zext(void) { diff --git a/kernel/bpf/inode.c b/kernel/bpf/inode.c index dd4b7fd60ee7..1576ff331ee4 100644 --- a/kernel/bpf/inode.c +++ b/kernel/bpf/inode.c @@ -122,7 +122,7 @@ static struct inode *bpf_get_inode(struct super_block *sb, inode->i_mtime = inode->i_atime; inode->i_ctime = inode->i_atime; - inode_init_owner(inode, dir, mode); + inode_init_owner(&init_user_ns, inode, dir, mode); return inode; } @@ -152,7 +152,8 @@ static void bpf_dentry_finalize(struct dentry *dentry, struct inode *inode, dir->i_ctime = dir->i_mtime; } -static int bpf_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode) +static int bpf_mkdir(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, umode_t mode) { struct inode *inode; @@ -381,8 +382,8 @@ bpf_lookup(struct inode *dir, struct dentry *dentry, unsigned flags) return simple_lookup(dir, dentry, flags); } -static int bpf_symlink(struct inode *dir, struct dentry *dentry, - const char *target) +static int bpf_symlink(struct user_namespace *mnt_userns, struct inode *dir, + struct dentry *dentry, const char *target) { char *link = kstrdup(target, GFP_USER | __GFP_NOWARN); struct inode *inode; @@ -507,7 +508,7 @@ static void *bpf_obj_do_get(const char __user *pathname, return ERR_PTR(ret); inode = d_backing_inode(path.dentry); - ret = inode_permission(inode, ACC_MODE(flags)); + ret = path_permission(&path, ACC_MODE(flags)); if (ret) goto out; @@ -558,7 +559,7 @@ int bpf_obj_get_user(const char __user *pathname, int flags) static struct bpf_prog *__get_prog_inode(struct inode *inode, enum bpf_prog_type type) { struct bpf_prog *prog; - int ret = inode_permission(inode, MAY_READ); + int ret = inode_permission(&init_user_ns, inode, MAY_READ); if (ret) return ERR_PTR(ret); diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c index 97eb0b2435b8..f9096b049cd6 100644 --- a/kernel/bpf/verifier.c +++ b/kernel/bpf/verifier.c @@ -531,6 +531,13 @@ static bool is_ptr_cast_function(enum bpf_func_id func_id) func_id == BPF_FUNC_skc_to_tcp_request_sock; } +static bool is_cmpxchg_insn(const struct bpf_insn *insn) +{ + return BPF_CLASS(insn->code) == BPF_STX && + BPF_MODE(insn->code) == BPF_ATOMIC && + insn->imm == BPF_CMPXCHG; +} + /* string representation of 'enum bpf_reg_type' */ static const char * const reg_type_str[] = { [NOT_INIT] = "?", @@ -1150,7 +1157,7 @@ static void mark_ptr_not_null_reg(struct bpf_reg_state *reg) reg->type = PTR_TO_RDWR_BUF; break; default: - WARN_ON("unknown nullable register type"); + WARN_ONCE(1, "unknown nullable register type"); } } @@ -1746,7 +1753,11 @@ static bool is_reg64(struct bpf_verifier_env *env, struct bpf_insn *insn, } if (class == BPF_STX) { - if (reg->type != SCALAR_VALUE) + /* BPF_STX (including atomic variants) has multiple source + * operands, one of which is a ptr. Check whether the caller is + * asking about it. + */ + if (t == SRC_OP && reg->type != SCALAR_VALUE) return true; return BPF_SIZE(code) == BPF_DW; } @@ -1778,22 +1789,38 @@ static bool is_reg64(struct bpf_verifier_env *env, struct bpf_insn *insn, return true; } -/* Return TRUE if INSN doesn't have explicit value define. */ -static bool insn_no_def(struct bpf_insn *insn) +/* Return the regno defined by the insn, or -1. */ +static int insn_def_regno(const struct bpf_insn *insn) { - u8 class = BPF_CLASS(insn->code); - - return (class == BPF_JMP || class == BPF_JMP32 || - class == BPF_STX || class == BPF_ST); + switch (BPF_CLASS(insn->code)) { + case BPF_JMP: + case BPF_JMP32: + case BPF_ST: + return -1; + case BPF_STX: + if (BPF_MODE(insn->code) == BPF_ATOMIC && + (insn->imm & BPF_FETCH)) { + if (insn->imm == BPF_CMPXCHG) + return BPF_REG_0; + else + return insn->src_reg; + } else { + return -1; + } + default: + return insn->dst_reg; + } } /* Return TRUE if INSN has defined any 32-bit value explicitly. */ static bool insn_has_def32(struct bpf_verifier_env *env, struct bpf_insn *insn) { - if (insn_no_def(insn)) + int dst_reg = insn_def_regno(insn); + + if (dst_reg == -1) return false; - return !is_reg64(env, insn, insn->dst_reg, NULL, DST_OP); + return !is_reg64(env, insn, dst_reg, NULL, DST_OP); } static void mark_insn_zext(struct bpf_verifier_env *env, @@ -11236,9 +11263,10 @@ static int opt_subreg_zext_lo32_rnd_hi32(struct bpf_verifier_env *env, for (i = 0; i < len; i++) { int adj_idx = i + delta; struct bpf_insn insn; - u8 load_reg; + int load_reg; insn = insns[adj_idx]; + load_reg = insn_def_regno(&insn); if (!aux[adj_idx].zext_dst) { u8 code, class; u32 imm_rnd; @@ -11248,14 +11276,14 @@ static int opt_subreg_zext_lo32_rnd_hi32(struct bpf_verifier_env *env, code = insn.code; class = BPF_CLASS(code); - if (insn_no_def(&insn)) + if (load_reg == -1) continue; /* NOTE: arg "reg" (the fourth one) is only used for - * BPF_STX which has been ruled out in above - * check, it is safe to pass NULL here. + * BPF_STX + SRC_OP, so it is safe to pass NULL + * here. */ - if (is_reg64(env, &insn, insn.dst_reg, NULL, DST_OP)) { + if (is_reg64(env, &insn, load_reg, NULL, DST_OP)) { if (class == BPF_LD && BPF_MODE(code) == BPF_IMM) i++; @@ -11270,31 +11298,28 @@ static int opt_subreg_zext_lo32_rnd_hi32(struct bpf_verifier_env *env, imm_rnd = get_random_int(); rnd_hi32_patch[0] = insn; rnd_hi32_patch[1].imm = imm_rnd; - rnd_hi32_patch[3].dst_reg = insn.dst_reg; + rnd_hi32_patch[3].dst_reg = load_reg; patch = rnd_hi32_patch; patch_len = 4; goto apply_patch_buffer; } - if (!bpf_jit_needs_zext()) + /* Add in an zero-extend instruction if a) the JIT has requested + * it or b) it's a CMPXCHG. + * + * The latter is because: BPF_CMPXCHG always loads a value into + * R0, therefore always zero-extends. However some archs' + * equivalent instruction only does this load when the + * comparison is successful. This detail of CMPXCHG is + * orthogonal to the general zero-extension behaviour of the + * CPU, so it's treated independently of bpf_jit_needs_zext. + */ + if (!bpf_jit_needs_zext() && !is_cmpxchg_insn(&insn)) continue; - /* zext_dst means that we want to zero-extend whatever register - * the insn defines, which is dst_reg most of the time, with - * the notable exception of BPF_STX + BPF_ATOMIC + BPF_FETCH. - */ - if (BPF_CLASS(insn.code) == BPF_STX && - BPF_MODE(insn.code) == BPF_ATOMIC) { - /* BPF_STX + BPF_ATOMIC insns without BPF_FETCH do not - * define any registers, therefore zext_dst cannot be - * set. - */ - if (WARN_ON(!(insn.imm & BPF_FETCH))) - return -EINVAL; - load_reg = insn.imm == BPF_CMPXCHG ? BPF_REG_0 - : insn.src_reg; - } else { - load_reg = insn.dst_reg; + if (WARN_ON(load_reg == -1)) { + verbose(env, "verifier bug. zext_dst is set, but no reg is defined\n"); + return -EFAULT; } zext_patch[0] = insn; diff --git a/kernel/capability.c b/kernel/capability.c index de7eac903a2a..46a361dde042 100644 --- a/kernel/capability.c +++ b/kernel/capability.c @@ -484,10 +484,12 @@ EXPORT_SYMBOL(file_ns_capable); * * Return true if the inode uid and gid are within the namespace. */ -bool privileged_wrt_inode_uidgid(struct user_namespace *ns, const struct inode *inode) +bool privileged_wrt_inode_uidgid(struct user_namespace *ns, + struct user_namespace *mnt_userns, + const struct inode *inode) { - return kuid_has_mapping(ns, inode->i_uid) && - kgid_has_mapping(ns, inode->i_gid); + return kuid_has_mapping(ns, i_uid_into_mnt(mnt_userns, inode)) && + kgid_has_mapping(ns, i_gid_into_mnt(mnt_userns, inode)); } /** @@ -499,11 +501,13 @@ bool privileged_wrt_inode_uidgid(struct user_namespace *ns, const struct inode * * its own user namespace and that the given inode's uid and gid are * mapped into the current user namespace. */ -bool capable_wrt_inode_uidgid(const struct inode *inode, int cap) +bool capable_wrt_inode_uidgid(struct user_namespace *mnt_userns, + const struct inode *inode, int cap) { struct user_namespace *ns = current_user_ns(); - return ns_capable(ns, cap) && privileged_wrt_inode_uidgid(ns, inode); + return ns_capable(ns, cap) && + privileged_wrt_inode_uidgid(ns, mnt_userns, inode); } EXPORT_SYMBOL(capable_wrt_inode_uidgid); diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c index 1ea995f801ec..9153b20e5cc6 100644 --- a/kernel/cgroup/cgroup.c +++ b/kernel/cgroup/cgroup.c @@ -4672,7 +4672,7 @@ static int cgroup_may_write(const struct cgroup *cgrp, struct super_block *sb) if (!inode) return -ENOMEM; - ret = inode_permission(inode, MAY_WRITE); + ret = inode_permission(&init_user_ns, inode, MAY_WRITE); iput(inode); return ret; } @@ -4728,8 +4728,8 @@ static int cgroup_attach_permissions(struct cgroup *src_cgrp, return ret; } -static ssize_t cgroup_procs_write(struct kernfs_open_file *of, - char *buf, size_t nbytes, loff_t off) +static ssize_t __cgroup_procs_write(struct kernfs_open_file *of, char *buf, + bool threadgroup) { struct cgroup *src_cgrp, *dst_cgrp; struct task_struct *task; @@ -4740,7 +4740,7 @@ static ssize_t cgroup_procs_write(struct kernfs_open_file *of, if (!dst_cgrp) return -ENODEV; - task = cgroup_procs_write_start(buf, true, &locked); + task = cgroup_procs_write_start(buf, threadgroup, &locked); ret = PTR_ERR_OR_ZERO(task); if (ret) goto out_unlock; @@ -4750,19 +4750,26 @@ static ssize_t cgroup_procs_write(struct kernfs_open_file *of, src_cgrp = task_cgroup_from_root(task, &cgrp_dfl_root); spin_unlock_irq(&css_set_lock); + /* process and thread migrations follow same delegation rule */ ret = cgroup_attach_permissions(src_cgrp, dst_cgrp, - of->file->f_path.dentry->d_sb, true); + of->file->f_path.dentry->d_sb, threadgroup); if (ret) goto out_finish; - ret = cgroup_attach_task(dst_cgrp, task, true); + ret = cgroup_attach_task(dst_cgrp, task, threadgroup); out_finish: cgroup_procs_write_finish(task, locked); out_unlock: cgroup_kn_unlock(of->kn); - return ret ?: nbytes; + return ret; +} + +static ssize_t cgroup_procs_write(struct kernfs_open_file *of, + char *buf, size_t nbytes, loff_t off) +{ + return __cgroup_procs_write(of, buf, true) ?: nbytes; } static void *cgroup_threads_start(struct seq_file *s, loff_t *pos) @@ -4773,41 +4780,7 @@ static void *cgroup_threads_start(struct seq_file *s, loff_t *pos) static ssize_t cgroup_threads_write(struct kernfs_open_file *of, char *buf, size_t nbytes, loff_t off) { - struct cgroup *src_cgrp, *dst_cgrp; - struct task_struct *task; - ssize_t ret; - bool locked; - - buf = strstrip(buf); - - dst_cgrp = cgroup_kn_lock_live(of->kn, false); - if (!dst_cgrp) - return -ENODEV; - - task = cgroup_procs_write_start(buf, false, &locked); - ret = PTR_ERR_OR_ZERO(task); - if (ret) - goto out_unlock; - - /* find the source cgroup */ - spin_lock_irq(&css_set_lock); - src_cgrp = task_cgroup_from_root(task, &cgrp_dfl_root); - spin_unlock_irq(&css_set_lock); - - /* thread migrations follow the cgroup.procs delegation rule */ - ret = cgroup_attach_permissions(src_cgrp, dst_cgrp, - of->file->f_path.dentry->d_sb, false); - if (ret) - goto out_finish; - - ret = cgroup_attach_task(dst_cgrp, task, false); - -out_finish: - cgroup_procs_write_finish(task, locked); -out_unlock: - cgroup_kn_unlock(of->kn); - - return ret ?: nbytes; + return __cgroup_procs_write(of, buf, false) ?: nbytes; } /* cgroup core interface files for the default hierarchy */ diff --git a/kernel/cgroup/cpuset.c b/kernel/cgroup/cpuset.c index 53c70c470a38..5258b68153e0 100644 --- a/kernel/cgroup/cpuset.c +++ b/kernel/cgroup/cpuset.c @@ -98,7 +98,7 @@ struct cpuset { * and if it ends up empty, it will inherit the parent's mask. * * - * On legacy hierachy: + * On legacy hierarchy: * * The user-configured masks are always the same with effective masks. */ @@ -1309,10 +1309,10 @@ static int update_parent_subparts_cpumask(struct cpuset *cpuset, int cmd, * @cs: the cpuset to consider * @tmp: temp variables for calculating effective_cpus & partition setup * - * When congifured cpumask is changed, the effective cpumasks of this cpuset + * When configured cpumask is changed, the effective cpumasks of this cpuset * and all its descendants need to be updated. * - * On legacy hierachy, effective_cpus will be the same with cpu_allowed. + * On legacy hierarchy, effective_cpus will be the same with cpu_allowed. * * Called with cpuset_mutex held */ diff --git a/kernel/debug/debug_core.c b/kernel/debug/debug_core.c index af6e8b4fb359..4708aec492df 100644 --- a/kernel/debug/debug_core.c +++ b/kernel/debug/debug_core.c @@ -119,7 +119,6 @@ static DEFINE_RAW_SPINLOCK(dbg_slave_lock); */ static atomic_t masters_in_kgdb; static atomic_t slaves_in_kgdb; -static atomic_t kgdb_break_tasklet_var; atomic_t kgdb_setting_breakpoint; struct task_struct *kgdb_usethread; @@ -456,6 +455,17 @@ setundefined: return 0; } +void kgdb_free_init_mem(void) +{ + int i; + + /* Clear init memory breakpoints. */ + for (i = 0; i < KGDB_MAX_BREAKPOINTS; i++) { + if (init_section_contains((void *)kgdb_break[i].bpt_addr, 0)) + kgdb_break[i].state = BP_UNDEFINED; + } +} + #ifdef CONFIG_KGDB_KDB void kdb_dump_stack_on_cpu(int cpu) { @@ -1084,31 +1094,6 @@ static void kgdb_unregister_callbacks(void) } } -/* - * There are times a tasklet needs to be used vs a compiled in - * break point so as to cause an exception outside a kgdb I/O module, - * such as is the case with kgdboe, where calling a breakpoint in the - * I/O driver itself would be fatal. - */ -static void kgdb_tasklet_bpt(unsigned long ing) -{ - kgdb_breakpoint(); - atomic_set(&kgdb_break_tasklet_var, 0); -} - -static DECLARE_TASKLET_OLD(kgdb_tasklet_breakpoint, kgdb_tasklet_bpt); - -void kgdb_schedule_breakpoint(void) -{ - if (atomic_read(&kgdb_break_tasklet_var) || - atomic_read(&kgdb_active) != -1 || - atomic_read(&kgdb_setting_breakpoint)) - return; - atomic_inc(&kgdb_break_tasklet_var); - tasklet_schedule(&kgdb_tasklet_breakpoint); -} -EXPORT_SYMBOL_GPL(kgdb_schedule_breakpoint); - /** * kgdb_register_io_module - register KGDB IO module * @new_dbg_io_ops: the io ops vector @@ -1166,7 +1151,7 @@ int kgdb_register_io_module(struct kgdb_io *new_dbg_io_ops) EXPORT_SYMBOL_GPL(kgdb_register_io_module); /** - * kkgdb_unregister_io_module - unregister KGDB IO module + * kgdb_unregister_io_module - unregister KGDB IO module * @old_dbg_io_ops: the io ops vector * * Unregister it with the KGDB core. diff --git a/kernel/debug/gdbstub.c b/kernel/debug/gdbstub.c index a77df59d9ca5..e149a0ac9e9e 100644 --- a/kernel/debug/gdbstub.c +++ b/kernel/debug/gdbstub.c @@ -595,7 +595,7 @@ static char *gdb_hex_reg_helper(int regnum, char *out) dbg_reg_def[i].size); } -/* Handle the 'p' individual regster get */ +/* Handle the 'p' individual register get */ static void gdb_cmd_reg_get(struct kgdb_state *ks) { unsigned long regnum; @@ -610,7 +610,7 @@ static void gdb_cmd_reg_get(struct kgdb_state *ks) gdb_hex_reg_helper(regnum, remcom_out_buffer); } -/* Handle the 'P' individual regster set */ +/* Handle the 'P' individual register set */ static void gdb_cmd_reg_set(struct kgdb_state *ks) { unsigned long regnum; diff --git a/kernel/debug/kdb/kdb_private.h b/kernel/debug/kdb/kdb_private.h index a4281fb99299..6cb92f7bbbd0 100644 --- a/kernel/debug/kdb/kdb_private.h +++ b/kernel/debug/kdb/kdb_private.h @@ -230,7 +230,7 @@ extern struct task_struct *kdb_curr_task(int); #define kdb_task_has_cpu(p) (task_curr(p)) -#define GFP_KDB (in_interrupt() ? GFP_ATOMIC : GFP_KERNEL) +#define GFP_KDB (in_dbg_master() ? GFP_ATOMIC : GFP_KERNEL) extern void *debug_kmalloc(size_t size, gfp_t flags); extern void debug_kfree(void *); @@ -254,4 +254,14 @@ extern char kdb_prompt_str[]; #define KDB_WORD_SIZE ((int)sizeof(unsigned long)) #endif /* CONFIG_KGDB_KDB */ + +#define kdb_func_printf(format, args...) \ + kdb_printf("%s: " format, __func__, ## args) + +#define kdb_dbg_printf(mask, format, args...) \ + do { \ + if (KDB_DEBUG(mask)) \ + kdb_func_printf(format, ## args); \ + } while (0) + #endif /* !_KDBPRIVATE_H */ diff --git a/kernel/debug/kdb/kdb_support.c b/kernel/debug/kdb/kdb_support.c index 6226502ce049..f7c1885abeb6 100644 --- a/kernel/debug/kdb/kdb_support.c +++ b/kernel/debug/kdb/kdb_support.c @@ -39,20 +39,15 @@ */ int kdbgetsymval(const char *symname, kdb_symtab_t *symtab) { - if (KDB_DEBUG(AR)) - kdb_printf("kdbgetsymval: symname=%s, symtab=%px\n", symname, - symtab); + kdb_dbg_printf(AR, "symname=%s, symtab=%px\n", symname, symtab); memset(symtab, 0, sizeof(*symtab)); symtab->sym_start = kallsyms_lookup_name(symname); if (symtab->sym_start) { - if (KDB_DEBUG(AR)) - kdb_printf("kdbgetsymval: returns 1, " - "symtab->sym_start=0x%lx\n", - symtab->sym_start); + kdb_dbg_printf(AR, "returns 1, symtab->sym_start=0x%lx\n", + symtab->sym_start); return 1; } - if (KDB_DEBUG(AR)) - kdb_printf("kdbgetsymval: returns 0\n"); + kdb_dbg_printf(AR, "returns 0\n"); return 0; } EXPORT_SYMBOL(kdbgetsymval); @@ -87,16 +82,14 @@ int kdbnearsym(unsigned long addr, kdb_symtab_t *symtab) #define knt1_size 128 /* must be >= kallsyms table size */ char *knt1 = NULL; - if (KDB_DEBUG(AR)) - kdb_printf("kdbnearsym: addr=0x%lx, symtab=%px\n", addr, symtab); + kdb_dbg_printf(AR, "addr=0x%lx, symtab=%px\n", addr, symtab); memset(symtab, 0, sizeof(*symtab)); if (addr < 4096) goto out; knt1 = debug_kmalloc(knt1_size, GFP_ATOMIC); if (!knt1) { - kdb_printf("kdbnearsym: addr=0x%lx cannot kmalloc knt1\n", - addr); + kdb_func_printf("addr=0x%lx cannot kmalloc knt1\n", addr); goto out; } symtab->sym_name = kallsyms_lookup(addr, &symbolsize , &offset, @@ -147,11 +140,8 @@ int kdbnearsym(unsigned long addr, kdb_symtab_t *symtab) if (symtab->mod_name == NULL) symtab->mod_name = "kernel"; - if (KDB_DEBUG(AR)) - kdb_printf("kdbnearsym: returns %d symtab->sym_start=0x%lx, " - "symtab->mod_name=%px, symtab->sym_name=%px (%s)\n", ret, - symtab->sym_start, symtab->mod_name, symtab->sym_name, - symtab->sym_name); + kdb_dbg_printf(AR, "returns %d symtab->sym_start=0x%lx, symtab->mod_name=%px, symtab->sym_name=%px (%s)\n", + ret, symtab->sym_start, symtab->mod_name, symtab->sym_name, symtab->sym_name); out: debug_kfree(knt1); @@ -328,7 +318,7 @@ int kdb_getarea_size(void *res, unsigned long addr, size_t size) int ret = copy_from_kernel_nofault((char *)res, (char *)addr, size); if (ret) { if (!KDB_STATE(SUPPRESS)) { - kdb_printf("kdb_getarea: Bad address 0x%lx\n", addr); + kdb_func_printf("Bad address 0x%lx\n", addr); KDB_STATE_SET(SUPPRESS); } ret = KDB_BADADDR; @@ -353,7 +343,7 @@ int kdb_putarea_size(unsigned long addr, void *res, size_t size) int ret = copy_from_kernel_nofault((char *)addr, (char *)res, size); if (ret) { if (!KDB_STATE(SUPPRESS)) { - kdb_printf("kdb_putarea: Bad address 0x%lx\n", addr); + kdb_func_printf("Bad address 0x%lx\n", addr); KDB_STATE_SET(SUPPRESS); } ret = KDB_BADADDR; @@ -435,7 +425,7 @@ int kdb_getphysword(unsigned long *word, unsigned long addr, size_t size) fallthrough; default: diag = KDB_BADWIDTH; - kdb_printf("kdb_getphysword: bad width %ld\n", (long) size); + kdb_func_printf("bad width %zu\n", size); } return diag; } @@ -484,7 +474,7 @@ int kdb_getword(unsigned long *word, unsigned long addr, size_t size) fallthrough; default: diag = KDB_BADWIDTH; - kdb_printf("kdb_getword: bad width %ld\n", (long) size); + kdb_func_printf("bad width %zu\n", size); } return diag; } @@ -528,7 +518,7 @@ int kdb_putword(unsigned long addr, unsigned long word, size_t size) fallthrough; default: diag = KDB_BADWIDTH; - kdb_printf("kdb_putword: bad width %ld\n", (long) size); + kdb_func_printf("bad width %zu\n", size); } return diag; } @@ -602,8 +592,7 @@ unsigned long kdb_task_state_string(const char *s) res = ~0UL; break; default: - kdb_printf("%s: unknown flag '%c' ignored\n", - __func__, *s); + kdb_func_printf("unknown flag '%c' ignored\n", *s); break; } ++s; @@ -884,18 +873,16 @@ void debug_kusage(void) if (!debug_kusage_one_time) goto out; debug_kusage_one_time = 0; - kdb_printf("%s: debug_kmalloc memory leak dah_first %d\n", - __func__, dah_first); + kdb_func_printf("debug_kmalloc memory leak dah_first %d\n", dah_first); if (dah_first) { h_used = (struct debug_alloc_header *)debug_alloc_pool; - kdb_printf("%s: h_used %px size %d\n", __func__, h_used, - h_used->size); + kdb_func_printf("h_used %px size %d\n", h_used, h_used->size); } do { h_used = (struct debug_alloc_header *) ((char *)h_free + dah_overhead + h_free->size); - kdb_printf("%s: h_used %px size %d caller %px\n", - __func__, h_used, h_used->size, h_used->caller); + kdb_func_printf("h_used %px size %d caller %px\n", + h_used, h_used->size, h_used->caller); h_free = (struct debug_alloc_header *) (debug_alloc_pool + h_free->next); } while (h_free->next); @@ -903,8 +890,8 @@ void debug_kusage(void) ((char *)h_free + dah_overhead + h_free->size); if ((char *)h_used - debug_alloc_pool != sizeof(debug_alloc_pool_aligned)) - kdb_printf("%s: h_used %px size %d caller %px\n", - __func__, h_used, h_used->size, h_used->caller); + kdb_func_printf("h_used %px size %d caller %px\n", + h_used, h_used->size, h_used->caller); out: spin_unlock(&dap_lock); } diff --git a/kernel/dma/Kconfig b/kernel/dma/Kconfig index 479fc145acfc..77b405508743 100644 --- a/kernel/dma/Kconfig +++ b/kernel/dma/Kconfig @@ -33,9 +33,6 @@ config NEED_DMA_MAP_STATE config ARCH_DMA_ADDR_T_64BIT def_bool 64BIT || PHYS_ADDR_T_64BIT -config ARCH_HAS_DMA_COHERENCE_H - bool - config ARCH_HAS_DMA_SET_MASK bool diff --git a/kernel/dma/map_benchmark.c b/kernel/dma/map_benchmark.c index da95df381483..e0e64f8b0739 100644 --- a/kernel/dma/map_benchmark.c +++ b/kernel/dma/map_benchmark.c @@ -21,6 +21,7 @@ #define DMA_MAP_BENCHMARK _IOWR('d', 1, struct map_benchmark) #define DMA_MAP_MAX_THREADS 1024 #define DMA_MAP_MAX_SECONDS 300 +#define DMA_MAP_MAX_TRANS_DELAY (10 * NSEC_PER_MSEC) #define DMA_MAP_BIDIRECTIONAL 0 #define DMA_MAP_TO_DEVICE 1 @@ -36,7 +37,8 @@ struct map_benchmark { __s32 node; /* which numa node this benchmark will run on */ __u32 dma_bits; /* DMA addressing capability */ __u32 dma_dir; /* DMA data direction */ - __u8 expansion[84]; /* For future use */ + __u32 dma_trans_ns; /* time for DMA transmission in ns */ + __u8 expansion[80]; /* For future use */ }; struct map_benchmark_data { @@ -87,6 +89,9 @@ static int map_benchmark_thread(void *data) map_etime = ktime_get(); map_delta = ktime_sub(map_etime, map_stime); + /* Pretend DMA is transmitting */ + ndelay(map->bparam.dma_trans_ns); + unmap_stime = ktime_get(); dma_unmap_single(map->dev, dma_addr, PAGE_SIZE, map->dir); unmap_etime = ktime_get(); @@ -218,6 +223,11 @@ static long map_benchmark_ioctl(struct file *file, unsigned int cmd, return -EINVAL; } + if (map->bparam.dma_trans_ns > DMA_MAP_MAX_TRANS_DELAY) { + pr_err("invalid transmission delay\n"); + return -EINVAL; + } + if (map->bparam.node != NUMA_NO_NODE && !node_possible(map->bparam.node)) { pr_err("invalid numa node\n"); diff --git a/kernel/dma/mapping.c b/kernel/dma/mapping.c index f87a89d08654..b6a633679933 100644 --- a/kernel/dma/mapping.c +++ b/kernel/dma/mapping.c @@ -16,6 +16,8 @@ #include "debug.h" #include "direct.h" +bool dma_default_coherent; + /* * Managed DMA API */ @@ -515,46 +517,6 @@ void dma_free_pages(struct device *dev, size_t size, struct page *page, } EXPORT_SYMBOL_GPL(dma_free_pages); -void *dma_alloc_noncoherent(struct device *dev, size_t size, - dma_addr_t *dma_handle, enum dma_data_direction dir, gfp_t gfp) -{ - const struct dma_map_ops *ops = get_dma_ops(dev); - void *vaddr; - - if (!ops || !ops->alloc_noncoherent) { - struct page *page; - - page = dma_alloc_pages(dev, size, dma_handle, dir, gfp); - if (!page) - return NULL; - return page_address(page); - } - - size = PAGE_ALIGN(size); - vaddr = ops->alloc_noncoherent(dev, size, dma_handle, dir, gfp); - if (vaddr) - debug_dma_map_page(dev, virt_to_page(vaddr), 0, size, dir, - *dma_handle); - return vaddr; -} -EXPORT_SYMBOL_GPL(dma_alloc_noncoherent); - -void dma_free_noncoherent(struct device *dev, size_t size, void *vaddr, - dma_addr_t dma_handle, enum dma_data_direction dir) -{ - const struct dma_map_ops *ops = get_dma_ops(dev); - - if (!ops || !ops->free_noncoherent) { - dma_free_pages(dev, size, virt_to_page(vaddr), dma_handle, dir); - return; - } - - size = PAGE_ALIGN(size); - debug_dma_unmap_page(dev, dma_handle, size, dir); - ops->free_noncoherent(dev, size, vaddr, dma_handle, dir); -} -EXPORT_SYMBOL_GPL(dma_free_noncoherent); - int dma_supported(struct device *dev, u64 mask) { const struct dma_map_ops *ops = get_dma_ops(dev); diff --git a/kernel/dma/swiotlb.c b/kernel/dma/swiotlb.c index 7c42df6e6100..c10e855a03bc 100644 --- a/kernel/dma/swiotlb.c +++ b/kernel/dma/swiotlb.c @@ -50,9 +50,6 @@ #define CREATE_TRACE_POINTS #include <trace/events/swiotlb.h> -#define OFFSET(val,align) ((unsigned long) \ - ( (val) & ( (align) - 1))) - #define SLABS_PER_PAGE (1 << (PAGE_SHIFT - IO_TLB_SHIFT)) /* @@ -103,6 +100,11 @@ static unsigned int max_segment; static phys_addr_t *io_tlb_orig_addr; /* + * The mapped buffer's size should be validated during a sync operation. + */ +static size_t *io_tlb_orig_size; + +/* * Protect the above data structures in the map and unmap calls */ static DEFINE_SPINLOCK(io_tlb_lock); @@ -171,7 +173,7 @@ void __init swiotlb_adjust_size(unsigned long new_size) * adjust/expand SWIOTLB size for their use. */ if (!io_tlb_nslabs) { - size = ALIGN(new_size, 1 << IO_TLB_SHIFT); + size = ALIGN(new_size, IO_TLB_SIZE); io_tlb_nslabs = size >> IO_TLB_SHIFT; io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE); @@ -192,6 +194,16 @@ void swiotlb_print_info(void) bytes >> 20); } +static inline unsigned long io_tlb_offset(unsigned long val) +{ + return val & (IO_TLB_SEGSIZE - 1); +} + +static inline unsigned long nr_slots(u64 val) +{ + return DIV_ROUND_UP(val, IO_TLB_SIZE); +} + /* * Early SWIOTLB allocation may be too early to allow an architecture to * perform the desired operations. This function allows the architecture to @@ -240,9 +252,16 @@ int __init swiotlb_init_with_tbl(char *tlb, unsigned long nslabs, int verbose) panic("%s: Failed to allocate %zu bytes align=0x%lx\n", __func__, alloc_size, PAGE_SIZE); + alloc_size = PAGE_ALIGN(io_tlb_nslabs * sizeof(size_t)); + io_tlb_orig_size = memblock_alloc(alloc_size, PAGE_SIZE); + if (!io_tlb_orig_size) + panic("%s: Failed to allocate %zu bytes align=0x%lx\n", + __func__, alloc_size, PAGE_SIZE); + for (i = 0; i < io_tlb_nslabs; i++) { - io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE); + io_tlb_list[i] = IO_TLB_SEGSIZE - io_tlb_offset(i); io_tlb_orig_addr[i] = INVALID_PHYS_ADDR; + io_tlb_orig_size[i] = 0; } io_tlb_index = 0; no_iotlb_memory = false; @@ -363,7 +382,7 @@ swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs) * between io_tlb_start and io_tlb_end. */ io_tlb_list = (unsigned int *)__get_free_pages(GFP_KERNEL, - get_order(io_tlb_nslabs * sizeof(int))); + get_order(io_tlb_nslabs * sizeof(int))); if (!io_tlb_list) goto cleanup3; @@ -374,9 +393,18 @@ swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs) if (!io_tlb_orig_addr) goto cleanup4; + io_tlb_orig_size = (size_t *) + __get_free_pages(GFP_KERNEL, + get_order(io_tlb_nslabs * + sizeof(size_t))); + if (!io_tlb_orig_size) + goto cleanup5; + + for (i = 0; i < io_tlb_nslabs; i++) { - io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE); + io_tlb_list[i] = IO_TLB_SEGSIZE - io_tlb_offset(i); io_tlb_orig_addr[i] = INVALID_PHYS_ADDR; + io_tlb_orig_size[i] = 0; } io_tlb_index = 0; no_iotlb_memory = false; @@ -389,6 +417,10 @@ swiotlb_late_init_with_tbl(char *tlb, unsigned long nslabs) return 0; +cleanup5: + free_pages((unsigned long)io_tlb_orig_addr, get_order(io_tlb_nslabs * + sizeof(phys_addr_t))); + cleanup4: free_pages((unsigned long)io_tlb_list, get_order(io_tlb_nslabs * sizeof(int))); @@ -404,6 +436,8 @@ void __init swiotlb_exit(void) return; if (late_alloc) { + free_pages((unsigned long)io_tlb_orig_size, + get_order(io_tlb_nslabs * sizeof(size_t))); free_pages((unsigned long)io_tlb_orig_addr, get_order(io_tlb_nslabs * sizeof(phys_addr_t))); free_pages((unsigned long)io_tlb_list, get_order(io_tlb_nslabs * @@ -413,6 +447,8 @@ void __init swiotlb_exit(void) } else { memblock_free_late(__pa(io_tlb_orig_addr), PAGE_ALIGN(io_tlb_nslabs * sizeof(phys_addr_t))); + memblock_free_late(__pa(io_tlb_orig_size), + PAGE_ALIGN(io_tlb_nslabs * sizeof(size_t))); memblock_free_late(__pa(io_tlb_list), PAGE_ALIGN(io_tlb_nslabs * sizeof(int))); memblock_free_late(io_tlb_start, @@ -461,79 +497,71 @@ static void swiotlb_bounce(phys_addr_t orig_addr, phys_addr_t tlb_addr, } } -phys_addr_t swiotlb_tbl_map_single(struct device *hwdev, phys_addr_t orig_addr, - size_t mapping_size, size_t alloc_size, - enum dma_data_direction dir, unsigned long attrs) -{ - dma_addr_t tbl_dma_addr = phys_to_dma_unencrypted(hwdev, io_tlb_start); - unsigned long flags; - phys_addr_t tlb_addr; - unsigned int nslots, stride, index, wrap; - int i; - unsigned long mask; - unsigned long offset_slots; - unsigned long max_slots; - unsigned long tmp_io_tlb_used; - - if (no_iotlb_memory) - panic("Can not allocate SWIOTLB buffer earlier and can't now provide you with the DMA bounce buffer"); - - if (mem_encrypt_active()) - pr_warn_once("Memory encryption is active and system is using DMA bounce buffers\n"); +#define slot_addr(start, idx) ((start) + ((idx) << IO_TLB_SHIFT)) - if (mapping_size > alloc_size) { - dev_warn_once(hwdev, "Invalid sizes (mapping: %zd bytes, alloc: %zd bytes)", - mapping_size, alloc_size); - return (phys_addr_t)DMA_MAPPING_ERROR; - } +/* + * Return the offset into a iotlb slot required to keep the device happy. + */ +static unsigned int swiotlb_align_offset(struct device *dev, u64 addr) +{ + return addr & dma_get_min_align_mask(dev) & (IO_TLB_SIZE - 1); +} - mask = dma_get_seg_boundary(hwdev); +/* + * Carefully handle integer overflow which can occur when boundary_mask == ~0UL. + */ +static inline unsigned long get_max_slots(unsigned long boundary_mask) +{ + if (boundary_mask == ~0UL) + return 1UL << (BITS_PER_LONG - IO_TLB_SHIFT); + return nr_slots(boundary_mask + 1); +} - tbl_dma_addr &= mask; +static unsigned int wrap_index(unsigned int index) +{ + if (index >= io_tlb_nslabs) + return 0; + return index; +} - offset_slots = ALIGN(tbl_dma_addr, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT; +/* + * Find a suitable number of IO TLB entries size that will fit this request and + * allocate a buffer from that IO TLB pool. + */ +static int find_slots(struct device *dev, phys_addr_t orig_addr, + size_t alloc_size) +{ + unsigned long boundary_mask = dma_get_seg_boundary(dev); + dma_addr_t tbl_dma_addr = + phys_to_dma_unencrypted(dev, io_tlb_start) & boundary_mask; + unsigned long max_slots = get_max_slots(boundary_mask); + unsigned int iotlb_align_mask = + dma_get_min_align_mask(dev) & ~(IO_TLB_SIZE - 1); + unsigned int nslots = nr_slots(alloc_size), stride; + unsigned int index, wrap, count = 0, i; + unsigned long flags; - /* - * Carefully handle integer overflow which can occur when mask == ~0UL. - */ - max_slots = mask + 1 - ? ALIGN(mask + 1, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT - : 1UL << (BITS_PER_LONG - IO_TLB_SHIFT); + BUG_ON(!nslots); /* - * For mappings greater than or equal to a page, we limit the stride - * (and hence alignment) to a page size. + * For mappings with an alignment requirement don't bother looping to + * unaligned slots once we found an aligned one. For allocations of + * PAGE_SIZE or larger only look for page aligned allocations. */ - nslots = ALIGN(alloc_size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT; + stride = (iotlb_align_mask >> IO_TLB_SHIFT) + 1; if (alloc_size >= PAGE_SIZE) - stride = (1 << (PAGE_SHIFT - IO_TLB_SHIFT)); - else - stride = 1; - - BUG_ON(!nslots); + stride = max(stride, stride << (PAGE_SHIFT - IO_TLB_SHIFT)); - /* - * Find suitable number of IO TLB entries size that will fit this - * request and allocate a buffer from that IO TLB pool. - */ spin_lock_irqsave(&io_tlb_lock, flags); - if (unlikely(nslots > io_tlb_nslabs - io_tlb_used)) goto not_found; - index = ALIGN(io_tlb_index, stride); - if (index >= io_tlb_nslabs) - index = 0; - wrap = index; - + index = wrap = wrap_index(ALIGN(io_tlb_index, stride)); do { - while (iommu_is_span_boundary(index, nslots, offset_slots, - max_slots)) { - index += stride; - if (index >= io_tlb_nslabs) - index = 0; - if (index == wrap) - goto not_found; + if ((slot_addr(tbl_dma_addr, index) & iotlb_align_mask) != + (orig_addr & iotlb_align_mask)) { + index = wrap_index(index + 1); + continue; } /* @@ -541,55 +569,96 @@ phys_addr_t swiotlb_tbl_map_single(struct device *hwdev, phys_addr_t orig_addr, * contiguous buffers, we allocate the buffers from that slot * and mark the entries as '0' indicating unavailable. */ - if (io_tlb_list[index] >= nslots) { - int count = 0; - - for (i = index; i < (int) (index + nslots); i++) - io_tlb_list[i] = 0; - for (i = index - 1; (OFFSET(i, IO_TLB_SEGSIZE) != IO_TLB_SEGSIZE - 1) && io_tlb_list[i]; i--) - io_tlb_list[i] = ++count; - tlb_addr = io_tlb_start + (index << IO_TLB_SHIFT); - - /* - * Update the indices to avoid searching in the next - * round. - */ - io_tlb_index = ((index + nslots) < io_tlb_nslabs - ? (index + nslots) : 0); - - goto found; + if (!iommu_is_span_boundary(index, nslots, + nr_slots(tbl_dma_addr), + max_slots)) { + if (io_tlb_list[index] >= nslots) + goto found; } - index += stride; - if (index >= io_tlb_nslabs) - index = 0; + index = wrap_index(index + stride); } while (index != wrap); not_found: - tmp_io_tlb_used = io_tlb_used; - spin_unlock_irqrestore(&io_tlb_lock, flags); - if (!(attrs & DMA_ATTR_NO_WARN) && printk_ratelimit()) - dev_warn(hwdev, "swiotlb buffer is full (sz: %zd bytes), total %lu (slots), used %lu (slots)\n", - alloc_size, io_tlb_nslabs, tmp_io_tlb_used); - return (phys_addr_t)DMA_MAPPING_ERROR; + return -1; + found: + for (i = index; i < index + nslots; i++) + io_tlb_list[i] = 0; + for (i = index - 1; + io_tlb_offset(i) != IO_TLB_SEGSIZE - 1 && + io_tlb_list[i]; i--) + io_tlb_list[i] = ++count; + + /* + * Update the indices to avoid searching in the next round. + */ + if (index + nslots < io_tlb_nslabs) + io_tlb_index = index + nslots; + else + io_tlb_index = 0; io_tlb_used += nslots; + spin_unlock_irqrestore(&io_tlb_lock, flags); + return index; +} + +phys_addr_t swiotlb_tbl_map_single(struct device *dev, phys_addr_t orig_addr, + size_t mapping_size, size_t alloc_size, + enum dma_data_direction dir, unsigned long attrs) +{ + unsigned int offset = swiotlb_align_offset(dev, orig_addr); + unsigned int index, i; + phys_addr_t tlb_addr; + + if (no_iotlb_memory) + panic("Can not allocate SWIOTLB buffer earlier and can't now provide you with the DMA bounce buffer"); + + if (mem_encrypt_active()) + pr_warn_once("Memory encryption is active and system is using DMA bounce buffers\n"); + + if (mapping_size > alloc_size) { + dev_warn_once(dev, "Invalid sizes (mapping: %zd bytes, alloc: %zd bytes)", + mapping_size, alloc_size); + return (phys_addr_t)DMA_MAPPING_ERROR; + } + + index = find_slots(dev, orig_addr, alloc_size + offset); + if (index == -1) { + if (!(attrs & DMA_ATTR_NO_WARN)) + dev_warn_ratelimited(dev, + "swiotlb buffer is full (sz: %zd bytes), total %lu (slots), used %lu (slots)\n", + alloc_size, io_tlb_nslabs, io_tlb_used); + return (phys_addr_t)DMA_MAPPING_ERROR; + } /* * Save away the mapping from the original address to the DMA address. * This is needed when we sync the memory. Then we sync the buffer if * needed. */ - for (i = 0; i < nslots; i++) - io_tlb_orig_addr[index+i] = orig_addr + (i << IO_TLB_SHIFT); + for (i = 0; i < nr_slots(alloc_size + offset); i++) { + io_tlb_orig_addr[index + i] = slot_addr(orig_addr, i); + io_tlb_orig_size[index+i] = alloc_size - (i << IO_TLB_SHIFT); + } + tlb_addr = slot_addr(io_tlb_start, index) + offset; if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC) && (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)) swiotlb_bounce(orig_addr, tlb_addr, mapping_size, DMA_TO_DEVICE); - return tlb_addr; } +static void validate_sync_size_and_truncate(struct device *hwdev, size_t orig_size, size_t *size) +{ + if (*size > orig_size) { + /* Warn and truncate mapping_size */ + dev_WARN_ONCE(hwdev, 1, + "Attempt for buffer overflow. Original size: %zu. Mapping size: %zu.\n", + orig_size, *size); + *size = orig_size; + } +} + /* * tlb_addr is the physical address of the bounce buffer to unmap. */ @@ -598,10 +667,13 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr, enum dma_data_direction dir, unsigned long attrs) { unsigned long flags; - int i, count, nslots = ALIGN(alloc_size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT; - int index = (tlb_addr - io_tlb_start) >> IO_TLB_SHIFT; + unsigned int offset = swiotlb_align_offset(hwdev, tlb_addr); + int i, count, nslots = nr_slots(alloc_size + offset); + int index = (tlb_addr - offset - io_tlb_start) >> IO_TLB_SHIFT; phys_addr_t orig_addr = io_tlb_orig_addr[index]; + validate_sync_size_and_truncate(hwdev, io_tlb_orig_size[index], &mapping_size); + /* * First, sync the memory before unmapping the entry */ @@ -617,26 +689,30 @@ void swiotlb_tbl_unmap_single(struct device *hwdev, phys_addr_t tlb_addr, * with slots below and above the pool being returned. */ spin_lock_irqsave(&io_tlb_lock, flags); - { - count = ((index + nslots) < ALIGN(index + 1, IO_TLB_SEGSIZE) ? - io_tlb_list[index + nslots] : 0); - /* - * Step 1: return the slots to the free list, merging the - * slots with superceeding slots - */ - for (i = index + nslots - 1; i >= index; i--) { - io_tlb_list[i] = ++count; - io_tlb_orig_addr[i] = INVALID_PHYS_ADDR; - } - /* - * Step 2: merge the returned slots with the preceding slots, - * if available (non zero) - */ - for (i = index - 1; (OFFSET(i, IO_TLB_SEGSIZE) != IO_TLB_SEGSIZE -1) && io_tlb_list[i]; i--) - io_tlb_list[i] = ++count; + if (index + nslots < ALIGN(index + 1, IO_TLB_SEGSIZE)) + count = io_tlb_list[index + nslots]; + else + count = 0; - io_tlb_used -= nslots; + /* + * Step 1: return the slots to the free list, merging the slots with + * superceeding slots + */ + for (i = index + nslots - 1; i >= index; i--) { + io_tlb_list[i] = ++count; + io_tlb_orig_addr[i] = INVALID_PHYS_ADDR; + io_tlb_orig_size[i] = 0; } + + /* + * Step 2: merge the returned slots with the preceding slots, if + * available (non zero) + */ + for (i = index - 1; + io_tlb_offset(i) != IO_TLB_SEGSIZE - 1 && io_tlb_list[i]; + i--) + io_tlb_list[i] = ++count; + io_tlb_used -= nslots; spin_unlock_irqrestore(&io_tlb_lock, flags); } @@ -645,11 +721,13 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr, enum dma_sync_target target) { int index = (tlb_addr - io_tlb_start) >> IO_TLB_SHIFT; + size_t orig_size = io_tlb_orig_size[index]; phys_addr_t orig_addr = io_tlb_orig_addr[index]; if (orig_addr == INVALID_PHYS_ADDR) return; - orig_addr += (unsigned long)tlb_addr & ((1 << IO_TLB_SHIFT) - 1); + + validate_sync_size_and_truncate(hwdev, orig_size, &size); switch (target) { case SYNC_FOR_CPU: @@ -707,7 +785,7 @@ dma_addr_t swiotlb_map(struct device *dev, phys_addr_t paddr, size_t size, size_t swiotlb_max_mapping_size(struct device *dev) { - return ((size_t)1 << IO_TLB_SHIFT) * IO_TLB_SEGSIZE; + return ((size_t)IO_TLB_SIZE) * IO_TLB_SEGSIZE; } bool is_swiotlb_active(void) diff --git a/kernel/events/core.c b/kernel/events/core.c index 129dee540a8b..0aeca5f3c0ac 100644 --- a/kernel/events/core.c +++ b/kernel/events/core.c @@ -269,7 +269,7 @@ static void event_function_call(struct perf_event *event, event_f func, void *da if (!event->parent) { /* * If this is a !child event, we must hold ctx::mutex to - * stabilize the the event->ctx relation. See + * stabilize the event->ctx relation. See * perf_event_ctx_lock(). */ lockdep_assert_held(&ctx->mutex); @@ -1303,7 +1303,7 @@ static void put_ctx(struct perf_event_context *ctx) * life-time rules separate them. That is an exiting task cannot fork, and a * spawning task cannot (yet) exit. * - * But remember that that these are parent<->child context relations, and + * But remember that these are parent<->child context relations, and * migration does not affect children, therefore these two orderings should not * interact. * @@ -1442,7 +1442,7 @@ static u64 primary_event_id(struct perf_event *event) /* * Get the perf_event_context for a task and lock it. * - * This has to cope with with the fact that until it is locked, + * This has to cope with the fact that until it is locked, * the context could get moved to another task. */ static struct perf_event_context * @@ -2486,7 +2486,7 @@ static void perf_set_shadow_time(struct perf_event *event, * But this is a bit hairy. * * So instead, we have an explicit cgroup call to remain - * within the time time source all along. We believe it + * within the time source all along. We believe it * is cleaner and simpler to understand. */ if (is_cgroup_event(event)) diff --git a/kernel/events/uprobes.c b/kernel/events/uprobes.c index 3ea7f8f92f1d..6addc9780319 100644 --- a/kernel/events/uprobes.c +++ b/kernel/events/uprobes.c @@ -1733,7 +1733,7 @@ void uprobe_free_utask(struct task_struct *t) } /* - * Allocate a uprobe_task object for the task if if necessary. + * Allocate a uprobe_task object for the task if necessary. * Called when the thread hits a breakpoint. * * Returns: diff --git a/kernel/fork.c b/kernel/fork.c index 181604db2d65..b94391a58708 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1942,6 +1942,8 @@ static __latent_entropy struct task_struct *copy_process( p = dup_task_struct(current, node); if (!p) goto fork_out; + if (args->io_thread) + p->flags |= PF_IO_WORKER; /* * This _must_ happen before we call free_task(), i.e. before we jump @@ -2416,6 +2418,34 @@ struct mm_struct *copy_init_mm(void) } /* + * This is like kernel_clone(), but shaved down and tailored to just + * creating io_uring workers. It returns a created task, or an error pointer. + * The returned task is inactive, and the caller must fire it up through + * wake_up_new_task(p). All signals are blocked in the created task. + */ +struct task_struct *create_io_thread(int (*fn)(void *), void *arg, int node) +{ + unsigned long flags = CLONE_FS|CLONE_FILES|CLONE_SIGHAND|CLONE_THREAD| + CLONE_IO; + struct kernel_clone_args args = { + .flags = ((lower_32_bits(flags) | CLONE_VM | + CLONE_UNTRACED) & ~CSIGNAL), + .exit_signal = (lower_32_bits(flags) & CSIGNAL), + .stack = (unsigned long)fn, + .stack_size = (unsigned long)arg, + .io_thread = 1, + }; + struct task_struct *tsk; + + tsk = copy_process(NULL, 0, node, &args); + if (!IS_ERR(tsk)) { + sigfillset(&tsk->blocked); + sigdelsetmask(&tsk->blocked, sigmask(SIGKILL)); + } + return tsk; +} + +/* * Ok, this is the main fork-routine. * * It copies the process, and if successful kick-starts diff --git a/kernel/groups.c b/kernel/groups.c index fe7e6385530e..787b381c7c00 100644 --- a/kernel/groups.c +++ b/kernel/groups.c @@ -15,12 +15,7 @@ struct group_info *groups_alloc(int gidsetsize) { struct group_info *gi; - unsigned int len; - - len = sizeof(struct group_info) + sizeof(kgid_t) * gidsetsize; - gi = kmalloc(len, GFP_KERNEL_ACCOUNT|__GFP_NOWARN|__GFP_NORETRY); - if (!gi) - gi = __vmalloc(len, GFP_KERNEL_ACCOUNT); + gi = kvmalloc(struct_size(gi, gid, gidsetsize), GFP_KERNEL_ACCOUNT); if (!gi) return NULL; diff --git a/kernel/irq/irqdomain.c b/kernel/irq/irqdomain.c index 6aacd342cd14..288151393a06 100644 --- a/kernel/irq/irqdomain.c +++ b/kernel/irq/irqdomain.c @@ -205,6 +205,7 @@ struct irq_domain *__irq_domain_add(struct fwnode_handle *fwnode, int size, } fwnode_handle_get(fwnode); + fwnode_dev_initialized(fwnode, true); /* Fill structure */ INIT_RADIX_TREE(&domain->revmap_tree, GFP_KERNEL); @@ -253,6 +254,7 @@ void irq_domain_remove(struct irq_domain *domain) pr_debug("Removed domain %s\n", domain->name); + fwnode_dev_initialized(domain->fwnode, false); fwnode_handle_put(domain->fwnode); if (domain->flags & IRQ_DOMAIN_NAME_ALLOCATED) kfree(domain->name); diff --git a/kernel/kallsyms.c b/kernel/kallsyms.c index fe9de067771c..8043a90aa50e 100644 --- a/kernel/kallsyms.c +++ b/kernel/kallsyms.c @@ -177,6 +177,11 @@ unsigned long kallsyms_lookup_name(const char *name) return module_kallsyms_lookup_name(name); } +#ifdef CONFIG_LIVEPATCH +/* + * Iterate over all symbols in vmlinux. For symbols from modules use + * module_kallsyms_on_each_symbol instead. + */ int kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *, unsigned long), void *data) @@ -192,8 +197,9 @@ int kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *, if (ret != 0) return ret; } - return module_kallsyms_on_each_symbol(fn, data); + return 0; } +#endif /* CONFIG_LIVEPATCH */ static unsigned long get_symbol_pos(unsigned long addr, unsigned long *symbolsize, diff --git a/kernel/kexec_file.c b/kernel/kexec_file.c index b02086d70492..5c3447cf7ad5 100644 --- a/kernel/kexec_file.c +++ b/kernel/kexec_file.c @@ -166,6 +166,11 @@ void kimage_file_post_load_cleanup(struct kimage *image) vfree(pi->sechdrs); pi->sechdrs = NULL; +#ifdef CONFIG_IMA_KEXEC + vfree(image->ima_buffer); + image->ima_buffer = NULL; +#endif /* CONFIG_IMA_KEXEC */ + /* See if architecture has anything to cleanup post load */ arch_kimage_file_post_load_cleanup(image); diff --git a/kernel/kexec_internal.h b/kernel/kexec_internal.h index 39d30ccf8d87..48aaf2ac0d0d 100644 --- a/kernel/kexec_internal.h +++ b/kernel/kexec_internal.h @@ -13,8 +13,6 @@ void kimage_terminate(struct kimage *image); int kimage_is_destination_range(struct kimage *image, unsigned long start, unsigned long end); -int machine_kexec_post_load(struct kimage *image); - extern struct mutex kexec_mutex; #ifdef CONFIG_KEXEC_FILE diff --git a/kernel/kprobes.c b/kernel/kprobes.c index d5a3eb74a657..745f08fdd7a6 100644 --- a/kernel/kprobes.c +++ b/kernel/kprobes.c @@ -861,7 +861,6 @@ out: cpus_read_unlock(); } -#ifdef CONFIG_SYSCTL static void optimize_all_kprobes(void) { struct hlist_head *head; @@ -887,6 +886,7 @@ out: mutex_unlock(&kprobe_mutex); } +#ifdef CONFIG_SYSCTL static void unoptimize_all_kprobes(void) { struct hlist_head *head; @@ -1520,13 +1520,16 @@ valid: return ap; } -/* Return error if the kprobe is being re-registered */ -static inline int check_kprobe_rereg(struct kprobe *p) +/* + * Warn and return error if the kprobe is being re-registered since + * there must be a software bug. + */ +static inline int warn_kprobe_rereg(struct kprobe *p) { int ret = 0; mutex_lock(&kprobe_mutex); - if (__get_valid_kprobe(p)) + if (WARN_ON_ONCE(__get_valid_kprobe(p))) ret = -EINVAL; mutex_unlock(&kprobe_mutex); @@ -1614,7 +1617,7 @@ int register_kprobe(struct kprobe *p) return PTR_ERR(addr); p->addr = addr; - ret = check_kprobe_rereg(p); + ret = warn_kprobe_rereg(p); if (ret) return ret; @@ -1995,7 +1998,7 @@ int register_kretprobe(struct kretprobe *rp) return ret; /* If only rp->kp.addr is specified, check reregistering kprobes */ - if (rp->kp.addr && check_kprobe_rereg(&rp->kp)) + if (rp->kp.addr && warn_kprobe_rereg(&rp->kp)) return -EINVAL; if (kretprobe_blacklist_size) { @@ -2497,18 +2500,14 @@ static int __init init_kprobes(void) } } -#if defined(CONFIG_OPTPROBES) -#if defined(__ARCH_WANT_KPROBES_INSN_SLOT) - /* Init kprobe_optinsn_slots */ - kprobe_optinsn_slots.insn_size = MAX_OPTINSN_SIZE; -#endif - /* By default, kprobes can be optimized */ - kprobes_allow_optimization = true; -#endif - /* By default, kprobes are armed */ kprobes_all_disarmed = false; +#if defined(CONFIG_OPTPROBES) && defined(__ARCH_WANT_KPROBES_INSN_SLOT) + /* Init kprobe_optinsn_slots for allocation */ + kprobe_optinsn_slots.insn_size = MAX_OPTINSN_SIZE; +#endif + err = arch_init_kprobes(); if (!err) err = register_die_notifier(&kprobe_exceptions_nb); @@ -2523,6 +2522,21 @@ static int __init init_kprobes(void) } early_initcall(init_kprobes); +#if defined(CONFIG_OPTPROBES) +static int __init init_optprobes(void) +{ + /* + * Enable kprobe optimization - this kicks the optimizer which + * depends on synchronize_rcu_tasks() and ksoftirqd, that is + * not spawned in early initcall. So delay the optimization. + */ + optimize_all_kprobes(); + + return 0; +} +subsys_initcall(init_optprobes); +#endif + #ifdef CONFIG_DEBUG_FS static void report_probe(struct seq_file *pi, struct kprobe *p, const char *sym, int offset, char *modname, struct kprobe *pp) diff --git a/kernel/livepatch/core.c b/kernel/livepatch/core.c index f76fdb925532..335d988bd811 100644 --- a/kernel/livepatch/core.c +++ b/kernel/livepatch/core.c @@ -19,6 +19,7 @@ #include <linux/moduleloader.h> #include <linux/completion.h> #include <linux/memory.h> +#include <linux/rcupdate.h> #include <asm/cacheflush.h> #include "core.h" #include "patch.h" @@ -57,7 +58,7 @@ static void klp_find_object_module(struct klp_object *obj) if (!klp_is_module(obj)) return; - mutex_lock(&module_mutex); + rcu_read_lock_sched(); /* * We do not want to block removal of patched modules and therefore * we do not take a reference here. The patches are removed by @@ -74,7 +75,7 @@ static void klp_find_object_module(struct klp_object *obj) if (mod && mod->klp_alive) obj->mod = mod; - mutex_unlock(&module_mutex); + rcu_read_unlock_sched(); } static bool klp_initialized(void) @@ -163,12 +164,10 @@ static int klp_find_object_symbol(const char *objname, const char *name, .pos = sympos, }; - mutex_lock(&module_mutex); if (objname) module_kallsyms_on_each_symbol(klp_find_callback, &args); else kallsyms_on_each_symbol(klp_find_callback, &args); - mutex_unlock(&module_mutex); /* * Ensure an address was found. If sympos is 0, ensure symbol is unique; diff --git a/kernel/locking/qrwlock.c b/kernel/locking/qrwlock.c index fe9ca92faa2a..4786dd271b45 100644 --- a/kernel/locking/qrwlock.c +++ b/kernel/locking/qrwlock.c @@ -12,7 +12,6 @@ #include <linux/percpu.h> #include <linux/hardirq.h> #include <linux/spinlock.h> -#include <asm/qrwlock.h> /** * queued_read_lock_slowpath - acquire read lock of a queue rwlock diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c index 03b21135313c..48fff6437901 100644 --- a/kernel/locking/rtmutex.c +++ b/kernel/locking/rtmutex.c @@ -1420,7 +1420,7 @@ rt_mutex_fasttrylock(struct rt_mutex *lock, } /* - * Performs the wakeup of the the top-waiter and re-enables preemption. + * Performs the wakeup of the top-waiter and re-enables preemption. */ void rt_mutex_postunlock(struct wake_q_head *wake_q) { @@ -1819,7 +1819,7 @@ struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock) * been started. * @waiter: the pre-initialized rt_mutex_waiter * - * Wait for the the lock acquisition started on our behalf by + * Wait for the lock acquisition started on our behalf by * rt_mutex_start_proxy_lock(). Upon failure, the caller must call * rt_mutex_cleanup_proxy_lock(). * diff --git a/kernel/locking/rwsem.c b/kernel/locking/rwsem.c index ba67600c7b2c..abba5df50006 100644 --- a/kernel/locking/rwsem.c +++ b/kernel/locking/rwsem.c @@ -1048,7 +1048,7 @@ rwsem_down_write_slowpath(struct rw_semaphore *sem, int state) /* * If there were already threads queued before us and: - * 1) there are no no active locks, wake the front + * 1) there are no active locks, wake the front * queued process(es) as the handoff bit might be set. * 2) there are no active writers and some readers, the lock * must be read owned; so we try to wake any read lock diff --git a/kernel/locking/semaphore.c b/kernel/locking/semaphore.c index d9dd94defc0a..9aa855a96c4a 100644 --- a/kernel/locking/semaphore.c +++ b/kernel/locking/semaphore.c @@ -119,7 +119,7 @@ EXPORT_SYMBOL(down_killable); * @sem: the semaphore to be acquired * * Try to acquire the semaphore atomically. Returns 0 if the semaphore has - * been acquired successfully or 1 if it it cannot be acquired. + * been acquired successfully or 1 if it cannot be acquired. * * NOTE: This return value is inverted from both spin_trylock and * mutex_trylock! Be careful about this when converting code. diff --git a/kernel/module.c b/kernel/module.c index 4bf30e4b3eaa..30479355ab85 100644 --- a/kernel/module.c +++ b/kernel/module.c @@ -87,8 +87,7 @@ * 3) module_addr_min/module_addr_max. * (delete and add uses RCU list operations). */ -DEFINE_MUTEX(module_mutex); -EXPORT_SYMBOL_GPL(module_mutex); +static DEFINE_MUTEX(module_mutex); static LIST_HEAD(modules); /* Work queue for freeing init sections in success case */ @@ -256,11 +255,6 @@ static void mod_update_bounds(struct module *mod) struct list_head *kdb_modules = &modules; /* kdb needs the list of modules */ #endif /* CONFIG_KGDB_KDB */ -static void module_assert_mutex(void) -{ - lockdep_assert_held(&module_mutex); -} - static void module_assert_mutex_or_preempt(void) { #ifdef CONFIG_LOCKDEP @@ -414,19 +408,8 @@ extern const struct kernel_symbol __start___ksymtab[]; extern const struct kernel_symbol __stop___ksymtab[]; extern const struct kernel_symbol __start___ksymtab_gpl[]; extern const struct kernel_symbol __stop___ksymtab_gpl[]; -extern const struct kernel_symbol __start___ksymtab_gpl_future[]; -extern const struct kernel_symbol __stop___ksymtab_gpl_future[]; extern const s32 __start___kcrctab[]; extern const s32 __start___kcrctab_gpl[]; -extern const s32 __start___kcrctab_gpl_future[]; -#ifdef CONFIG_UNUSED_SYMBOLS -extern const struct kernel_symbol __start___ksymtab_unused[]; -extern const struct kernel_symbol __stop___ksymtab_unused[]; -extern const struct kernel_symbol __start___ksymtab_unused_gpl[]; -extern const struct kernel_symbol __stop___ksymtab_unused_gpl[]; -extern const s32 __start___kcrctab_unused[]; -extern const s32 __start___kcrctab_unused_gpl[]; -#endif #ifndef CONFIG_MODVERSIONS #define symversion(base, idx) NULL @@ -434,87 +417,14 @@ extern const s32 __start___kcrctab_unused_gpl[]; #define symversion(base, idx) ((base != NULL) ? ((base) + (idx)) : NULL) #endif -static bool each_symbol_in_section(const struct symsearch *arr, - unsigned int arrsize, - struct module *owner, - bool (*fn)(const struct symsearch *syms, - struct module *owner, - void *data), - void *data) -{ - unsigned int j; - - for (j = 0; j < arrsize; j++) { - if (fn(&arr[j], owner, data)) - return true; - } - - return false; -} - -/* Returns true as soon as fn returns true, otherwise false. */ -static bool each_symbol_section(bool (*fn)(const struct symsearch *arr, - struct module *owner, - void *data), - void *data) -{ - struct module *mod; - static const struct symsearch arr[] = { - { __start___ksymtab, __stop___ksymtab, __start___kcrctab, - NOT_GPL_ONLY, false }, - { __start___ksymtab_gpl, __stop___ksymtab_gpl, - __start___kcrctab_gpl, - GPL_ONLY, false }, - { __start___ksymtab_gpl_future, __stop___ksymtab_gpl_future, - __start___kcrctab_gpl_future, - WILL_BE_GPL_ONLY, false }, -#ifdef CONFIG_UNUSED_SYMBOLS - { __start___ksymtab_unused, __stop___ksymtab_unused, - __start___kcrctab_unused, - NOT_GPL_ONLY, true }, - { __start___ksymtab_unused_gpl, __stop___ksymtab_unused_gpl, - __start___kcrctab_unused_gpl, - GPL_ONLY, true }, -#endif - }; - - module_assert_mutex_or_preempt(); - - if (each_symbol_in_section(arr, ARRAY_SIZE(arr), NULL, fn, data)) - return true; - - list_for_each_entry_rcu(mod, &modules, list, - lockdep_is_held(&module_mutex)) { - struct symsearch arr[] = { - { mod->syms, mod->syms + mod->num_syms, mod->crcs, - NOT_GPL_ONLY, false }, - { mod->gpl_syms, mod->gpl_syms + mod->num_gpl_syms, - mod->gpl_crcs, - GPL_ONLY, false }, - { mod->gpl_future_syms, - mod->gpl_future_syms + mod->num_gpl_future_syms, - mod->gpl_future_crcs, - WILL_BE_GPL_ONLY, false }, -#ifdef CONFIG_UNUSED_SYMBOLS - { mod->unused_syms, - mod->unused_syms + mod->num_unused_syms, - mod->unused_crcs, - NOT_GPL_ONLY, true }, - { mod->unused_gpl_syms, - mod->unused_gpl_syms + mod->num_unused_gpl_syms, - mod->unused_gpl_crcs, - GPL_ONLY, true }, -#endif - }; - - if (mod->state == MODULE_STATE_UNFORMED) - continue; - - if (each_symbol_in_section(arr, ARRAY_SIZE(arr), mod, fn, data)) - return true; - } - return false; -} +struct symsearch { + const struct kernel_symbol *start, *stop; + const s32 *crcs; + enum mod_license { + NOT_GPL_ONLY, + GPL_ONLY, + } license; +}; struct find_symbol_arg { /* Input */ @@ -535,28 +445,8 @@ static bool check_exported_symbol(const struct symsearch *syms, { struct find_symbol_arg *fsa = data; - if (!fsa->gplok) { - if (syms->license == GPL_ONLY) - return false; - if (syms->license == WILL_BE_GPL_ONLY && fsa->warn) { - pr_warn("Symbol %s is being used by a non-GPL module, " - "which will not be allowed in the future\n", - fsa->name); - } - } - -#ifdef CONFIG_UNUSED_SYMBOLS - if (syms->unused && fsa->warn) { - pr_warn("Symbol %s is marked as UNUSED, however this module is " - "using it.\n", fsa->name); - pr_warn("This symbol will go away in the future.\n"); - pr_warn("Please evaluate if this is the right api to use and " - "if it really is, submit a report to the linux kernel " - "mailing list together with submitting your code for " - "inclusion.\n"); - } -#endif - + if (!fsa->gplok && syms->license == GPL_ONLY) + return false; fsa->owner = owner; fsa->crc = symversion(syms->crcs, symnum); fsa->sym = &syms->start[symnum]; @@ -619,31 +509,44 @@ static bool find_exported_symbol_in_section(const struct symsearch *syms, * Find an exported symbol and return it, along with, (optional) crc and * (optional) module which owns it. Needs preempt disabled or module_mutex. */ -static const struct kernel_symbol *find_symbol(const char *name, - struct module **owner, - const s32 **crc, - enum mod_license *license, - bool gplok, - bool warn) -{ - struct find_symbol_arg fsa; - - fsa.name = name; - fsa.gplok = gplok; - fsa.warn = warn; - - if (each_symbol_section(find_exported_symbol_in_section, &fsa)) { - if (owner) - *owner = fsa.owner; - if (crc) - *crc = fsa.crc; - if (license) - *license = fsa.license; - return fsa.sym; +static bool find_symbol(struct find_symbol_arg *fsa) +{ + static const struct symsearch arr[] = { + { __start___ksymtab, __stop___ksymtab, __start___kcrctab, + NOT_GPL_ONLY }, + { __start___ksymtab_gpl, __stop___ksymtab_gpl, + __start___kcrctab_gpl, + GPL_ONLY }, + }; + struct module *mod; + unsigned int i; + + module_assert_mutex_or_preempt(); + + for (i = 0; i < ARRAY_SIZE(arr); i++) + if (find_exported_symbol_in_section(&arr[i], NULL, fsa)) + return true; + + list_for_each_entry_rcu(mod, &modules, list, + lockdep_is_held(&module_mutex)) { + struct symsearch arr[] = { + { mod->syms, mod->syms + mod->num_syms, mod->crcs, + NOT_GPL_ONLY }, + { mod->gpl_syms, mod->gpl_syms + mod->num_gpl_syms, + mod->gpl_crcs, + GPL_ONLY }, + }; + + if (mod->state == MODULE_STATE_UNFORMED) + continue; + + for (i = 0; i < ARRAY_SIZE(arr); i++) + if (find_exported_symbol_in_section(&arr[i], mod, fsa)) + return true; } - pr_debug("Failed to find symbol %s\n", name); - return NULL; + pr_debug("Failed to find symbol %s\n", fsa->name); + return false; } /* @@ -669,10 +572,8 @@ static struct module *find_module_all(const char *name, size_t len, struct module *find_module(const char *name) { - module_assert_mutex(); return find_module_all(name, strlen(name), false); } -EXPORT_SYMBOL_GPL(find_module); #ifdef CONFIG_SMP @@ -1107,12 +1008,15 @@ static inline void print_unload_info(struct seq_file *m, struct module *mod) void __symbol_put(const char *symbol) { - struct module *owner; + struct find_symbol_arg fsa = { + .name = symbol, + .gplok = true, + }; preempt_disable(); - if (!find_symbol(symbol, &owner, NULL, NULL, true, false)) + if (!find_symbol(&fsa)) BUG(); - module_put(owner); + module_put(fsa.owner); preempt_enable(); } EXPORT_SYMBOL(__symbol_put); @@ -1381,19 +1285,22 @@ bad_version: static inline int check_modstruct_version(const struct load_info *info, struct module *mod) { - const s32 *crc; + struct find_symbol_arg fsa = { + .name = "module_layout", + .gplok = true, + }; /* * Since this should be found in kernel (which can't be removed), no * locking is necessary -- use preempt_disable() to placate lockdep. */ preempt_disable(); - if (!find_symbol("module_layout", NULL, &crc, NULL, true, false)) { + if (!find_symbol(&fsa)) { preempt_enable(); BUG(); } preempt_enable(); - return check_version(info, "module_layout", mod, crc); + return check_version(info, "module_layout", mod, fsa.crc); } /* First part is kernel version, which we ignore if module has crcs. */ @@ -1487,10 +1394,11 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod, const char *name, char ownername[]) { - struct module *owner; - const struct kernel_symbol *sym; - const s32 *crc; - enum mod_license license; + struct find_symbol_arg fsa = { + .name = name, + .gplok = !(mod->taints & (1 << TAINT_PROPRIETARY_MODULE)), + .warn = true, + }; int err; /* @@ -1500,42 +1408,40 @@ static const struct kernel_symbol *resolve_symbol(struct module *mod, */ sched_annotate_sleep(); mutex_lock(&module_mutex); - sym = find_symbol(name, &owner, &crc, &license, - !(mod->taints & (1 << TAINT_PROPRIETARY_MODULE)), true); - if (!sym) + if (!find_symbol(&fsa)) goto unlock; - if (license == GPL_ONLY) + if (fsa.license == GPL_ONLY) mod->using_gplonly_symbols = true; - if (!inherit_taint(mod, owner)) { - sym = NULL; + if (!inherit_taint(mod, fsa.owner)) { + fsa.sym = NULL; goto getname; } - if (!check_version(info, name, mod, crc)) { - sym = ERR_PTR(-EINVAL); + if (!check_version(info, name, mod, fsa.crc)) { + fsa.sym = ERR_PTR(-EINVAL); goto getname; } - err = verify_namespace_is_imported(info, sym, mod); + err = verify_namespace_is_imported(info, fsa.sym, mod); if (err) { - sym = ERR_PTR(err); + fsa.sym = ERR_PTR(err); goto getname; } - err = ref_module(mod, owner); + err = ref_module(mod, fsa.owner); if (err) { - sym = ERR_PTR(err); + fsa.sym = ERR_PTR(err); goto getname; } getname: /* We must make copy under the lock if we failed to get ref. */ - strncpy(ownername, module_name(owner), MODULE_NAME_LEN); + strncpy(ownername, module_name(fsa.owner), MODULE_NAME_LEN); unlock: mutex_unlock(&module_mutex); - return sym; + return fsa.sym; } static const struct kernel_symbol * @@ -2296,16 +2202,19 @@ static void free_module(struct module *mod) void *__symbol_get(const char *symbol) { - struct module *owner; - const struct kernel_symbol *sym; + struct find_symbol_arg fsa = { + .name = symbol, + .gplok = true, + .warn = true, + }; preempt_disable(); - sym = find_symbol(symbol, &owner, NULL, NULL, true, true); - if (sym && strong_try_module_get(owner)) - sym = NULL; + if (!find_symbol(&fsa) || strong_try_module_get(fsa.owner)) { + preempt_enable(); + return NULL; + } preempt_enable(); - - return sym ? (void *)kernel_symbol_value(sym) : NULL; + return (void *)kernel_symbol_value(fsa.sym); } EXPORT_SYMBOL_GPL(__symbol_get); @@ -2318,7 +2227,6 @@ EXPORT_SYMBOL_GPL(__symbol_get); static int verify_exported_symbols(struct module *mod) { unsigned int i; - struct module *owner; const struct kernel_symbol *s; struct { const struct kernel_symbol *sym; @@ -2326,21 +2234,19 @@ static int verify_exported_symbols(struct module *mod) } arr[] = { { mod->syms, mod->num_syms }, { mod->gpl_syms, mod->num_gpl_syms }, - { mod->gpl_future_syms, mod->num_gpl_future_syms }, -#ifdef CONFIG_UNUSED_SYMBOLS - { mod->unused_syms, mod->num_unused_syms }, - { mod->unused_gpl_syms, mod->num_unused_gpl_syms }, -#endif }; for (i = 0; i < ARRAY_SIZE(arr); i++) { for (s = arr[i].sym; s < arr[i].sym + arr[i].num; s++) { - if (find_symbol(kernel_symbol_name(s), &owner, NULL, - NULL, true, false)) { + struct find_symbol_arg fsa = { + .name = kernel_symbol_name(s), + .gplok = true, + }; + if (find_symbol(&fsa)) { pr_err("%s: exports duplicate symbol %s" " (owned by %s)\n", mod->name, kernel_symbol_name(s), - module_name(owner)); + module_name(fsa.owner)); return -ENOEXEC; } } @@ -2348,6 +2254,21 @@ static int verify_exported_symbols(struct module *mod) return 0; } +static bool ignore_undef_symbol(Elf_Half emachine, const char *name) +{ + /* + * On x86, PIC code and Clang non-PIC code may have call foo@PLT. GNU as + * before 2.37 produces an unreferenced _GLOBAL_OFFSET_TABLE_ on x86-64. + * i386 has a similar problem but may not deserve a fix. + * + * If we ever have to ignore many symbols, consider refactoring the code to + * only warn if referenced by a relocation. + */ + if (emachine == EM_386 || emachine == EM_X86_64) + return !strcmp(name, "_GLOBAL_OFFSET_TABLE_"); + return false; +} + /* Change all symbols so that st_value encodes the pointer directly. */ static int simplify_symbols(struct module *mod, const struct load_info *info) { @@ -2395,8 +2316,10 @@ static int simplify_symbols(struct module *mod, const struct load_info *info) break; } - /* Ok if weak. */ - if (!ksym && ELF_ST_BIND(sym[i].st_info) == STB_WEAK) + /* Ok if weak or ignored. */ + if (!ksym && + (ELF_ST_BIND(sym[i].st_info) == STB_WEAK || + ignore_undef_symbol(info->hdr->e_machine, name))) break; ret = PTR_ERR(ksym) ?: -ENOENT; @@ -2964,7 +2887,7 @@ static int module_sig_check(struct load_info *info, int flags) } if (is_module_sig_enforced()) { - pr_notice("%s: loading of %s is rejected\n", info->name, reason); + pr_notice("Loading of %s is rejected\n", reason); return -EKEYREJECTED; } @@ -2977,9 +2900,33 @@ static int module_sig_check(struct load_info *info, int flags) } #endif /* !CONFIG_MODULE_SIG */ -/* Sanity checks against invalid binaries, wrong arch, weird elf version. */ -static int elf_header_check(struct load_info *info) +static int validate_section_offset(struct load_info *info, Elf_Shdr *shdr) +{ + unsigned long secend; + + /* + * Check for both overflow and offset/size being + * too large. + */ + secend = shdr->sh_offset + shdr->sh_size; + if (secend < shdr->sh_offset || secend > info->len) + return -ENOEXEC; + + return 0; +} + +/* + * Sanity checks against invalid binaries, wrong arch, weird elf version. + * + * Also do basic validity checks against section offsets and sizes, the + * section name string table, and the indices used for it (sh_name). + */ +static int elf_validity_check(struct load_info *info) { + unsigned int i; + Elf_Shdr *shdr, *strhdr; + int err; + if (info->len < sizeof(*(info->hdr))) return -ENOEXEC; @@ -2989,11 +2936,78 @@ static int elf_header_check(struct load_info *info) || info->hdr->e_shentsize != sizeof(Elf_Shdr)) return -ENOEXEC; + /* + * e_shnum is 16 bits, and sizeof(Elf_Shdr) is + * known and small. So e_shnum * sizeof(Elf_Shdr) + * will not overflow unsigned long on any platform. + */ if (info->hdr->e_shoff >= info->len || (info->hdr->e_shnum * sizeof(Elf_Shdr) > info->len - info->hdr->e_shoff)) return -ENOEXEC; + info->sechdrs = (void *)info->hdr + info->hdr->e_shoff; + + /* + * Verify if the section name table index is valid. + */ + if (info->hdr->e_shstrndx == SHN_UNDEF + || info->hdr->e_shstrndx >= info->hdr->e_shnum) + return -ENOEXEC; + + strhdr = &info->sechdrs[info->hdr->e_shstrndx]; + err = validate_section_offset(info, strhdr); + if (err < 0) + return err; + + /* + * The section name table must be NUL-terminated, as required + * by the spec. This makes strcmp and pr_* calls that access + * strings in the section safe. + */ + info->secstrings = (void *)info->hdr + strhdr->sh_offset; + if (info->secstrings[strhdr->sh_size - 1] != '\0') + return -ENOEXEC; + + /* + * The code assumes that section 0 has a length of zero and + * an addr of zero, so check for it. + */ + if (info->sechdrs[0].sh_type != SHT_NULL + || info->sechdrs[0].sh_size != 0 + || info->sechdrs[0].sh_addr != 0) + return -ENOEXEC; + + for (i = 1; i < info->hdr->e_shnum; i++) { + shdr = &info->sechdrs[i]; + switch (shdr->sh_type) { + case SHT_NULL: + case SHT_NOBITS: + continue; + case SHT_SYMTAB: + if (shdr->sh_link == SHN_UNDEF + || shdr->sh_link >= info->hdr->e_shnum) + return -ENOEXEC; + fallthrough; + default: + err = validate_section_offset(info, shdr); + if (err < 0) { + pr_err("Invalid ELF section in module (section %u type %u)\n", + i, shdr->sh_type); + return err; + } + + if (shdr->sh_flags & SHF_ALLOC) { + if (shdr->sh_name >= strhdr->sh_size) { + pr_err("Invalid ELF section name in module (section %u type %u)\n", + i, shdr->sh_type); + return -ENOEXEC; + } + } + break; + } + } + return 0; } @@ -3095,11 +3109,6 @@ static int rewrite_section_headers(struct load_info *info, int flags) for (i = 1; i < info->hdr->e_shnum; i++) { Elf_Shdr *shdr = &info->sechdrs[i]; - if (shdr->sh_type != SHT_NOBITS - && info->len < shdr->sh_offset + shdr->sh_size) { - pr_err("Module len %lu truncated\n", info->len); - return -ENOEXEC; - } /* * Mark all sections sh_addr with their address in the @@ -3133,11 +3142,6 @@ static int setup_load_info(struct load_info *info, int flags) { unsigned int i; - /* Set up the convenience variables */ - info->sechdrs = (void *)info->hdr + info->hdr->e_shoff; - info->secstrings = (void *)info->hdr - + info->sechdrs[info->hdr->e_shstrndx].sh_offset; - /* Try to find a name early so we can log errors with a module name */ info->index.info = find_sec(info, ".modinfo"); if (info->index.info) @@ -3241,22 +3245,7 @@ static int find_module_sections(struct module *mod, struct load_info *info) sizeof(*mod->gpl_syms), &mod->num_gpl_syms); mod->gpl_crcs = section_addr(info, "__kcrctab_gpl"); - mod->gpl_future_syms = section_objs(info, - "__ksymtab_gpl_future", - sizeof(*mod->gpl_future_syms), - &mod->num_gpl_future_syms); - mod->gpl_future_crcs = section_addr(info, "__kcrctab_gpl_future"); - -#ifdef CONFIG_UNUSED_SYMBOLS - mod->unused_syms = section_objs(info, "__ksymtab_unused", - sizeof(*mod->unused_syms), - &mod->num_unused_syms); - mod->unused_crcs = section_addr(info, "__kcrctab_unused"); - mod->unused_gpl_syms = section_objs(info, "__ksymtab_unused_gpl", - sizeof(*mod->unused_gpl_syms), - &mod->num_unused_gpl_syms); - mod->unused_gpl_crcs = section_addr(info, "__kcrctab_unused_gpl"); -#endif + #ifdef CONFIG_CONSTRUCTORS mod->ctors = section_objs(info, ".ctors", sizeof(*mod->ctors), &mod->num_ctors); @@ -3437,14 +3426,8 @@ static int check_module_license_and_versions(struct module *mod) pr_warn("%s: module license taints kernel.\n", mod->name); #ifdef CONFIG_MODVERSIONS - if ((mod->num_syms && !mod->crcs) - || (mod->num_gpl_syms && !mod->gpl_crcs) - || (mod->num_gpl_future_syms && !mod->gpl_future_crcs) -#ifdef CONFIG_UNUSED_SYMBOLS - || (mod->num_unused_syms && !mod->unused_crcs) - || (mod->num_unused_gpl_syms && !mod->unused_gpl_crcs) -#endif - ) { + if ((mod->num_syms && !mod->crcs) || + (mod->num_gpl_syms && !mod->gpl_crcs)) { return try_to_force_load(mod, "no versions for exported symbols"); } @@ -3894,26 +3877,50 @@ static int load_module(struct load_info *info, const char __user *uargs, long err = 0; char *after_dashes; - err = elf_header_check(info); + /* + * Do the signature check (if any) first. All that + * the signature check needs is info->len, it does + * not need any of the section info. That can be + * set up later. This will minimize the chances + * of a corrupt module causing problems before + * we even get to the signature check. + * + * The check will also adjust info->len by stripping + * off the sig length at the end of the module, making + * checks against info->len more correct. + */ + err = module_sig_check(info, flags); + if (err) + goto free_copy; + + /* + * Do basic sanity checks against the ELF header and + * sections. + */ + err = elf_validity_check(info); if (err) { - pr_err("Module has invalid ELF header\n"); + pr_err("Module has invalid ELF structures\n"); goto free_copy; } + /* + * Everything checks out, so set up the section info + * in the info structure. + */ err = setup_load_info(info, flags); if (err) goto free_copy; + /* + * Now that we know we have the correct module name, check + * if it's blacklisted. + */ if (blacklisted(info->name)) { err = -EPERM; pr_err("Module %s is blacklisted\n", info->name); goto free_copy; } - err = module_sig_check(info, flags); - if (err) - goto free_copy; - err = rewrite_section_headers(info, flags); if (err) goto free_copy; @@ -4374,16 +4381,16 @@ unsigned long module_kallsyms_lookup_name(const char *name) return ret; } +#ifdef CONFIG_LIVEPATCH int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, struct module *, unsigned long), void *data) { struct module *mod; unsigned int i; - int ret; - - module_assert_mutex(); + int ret = 0; + mutex_lock(&module_mutex); list_for_each_entry(mod, &modules, list) { /* We hold module_mutex: no need for rcu_dereference_sched */ struct mod_kallsyms *kallsyms = mod->kallsyms; @@ -4399,11 +4406,13 @@ int module_kallsyms_on_each_symbol(int (*fn)(void *, const char *, ret = fn(data, kallsyms_symbol_name(kallsyms, i), mod, kallsyms_symbol_value(sym)); if (ret != 0) - return ret; + break; } } - return 0; + mutex_unlock(&module_mutex); + return ret; } +#endif /* CONFIG_LIVEPATCH */ #endif /* CONFIG_KALLSYMS */ /* Maximum number of characters written by module_flags() */ diff --git a/kernel/module_signature.c b/kernel/module_signature.c index 4224a1086b7d..00132d12487c 100644 --- a/kernel/module_signature.c +++ b/kernel/module_signature.c @@ -25,7 +25,7 @@ int mod_check_sig(const struct module_signature *ms, size_t file_len, return -EBADMSG; if (ms->id_type != PKEY_ID_PKCS7) { - pr_err("%s: Module is not signed with expected PKCS#7 message\n", + pr_err("%s: not signed with expected PKCS#7 message\n", name); return -ENOPKG; } diff --git a/kernel/module_signing.c b/kernel/module_signing.c index 9d9fc678c91d..8723ae70ea1f 100644 --- a/kernel/module_signing.c +++ b/kernel/module_signing.c @@ -30,7 +30,7 @@ int mod_verify_sig(const void *mod, struct load_info *info) memcpy(&ms, mod + (modlen - sizeof(ms)), sizeof(ms)); - ret = mod_check_sig(&ms, modlen, info->name); + ret = mod_check_sig(&ms, modlen, "module"); if (ret) return ret; diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c index 5a95c688621f..575a34b88936 100644 --- a/kernel/printk/printk.c +++ b/kernel/printk/printk.c @@ -735,9 +735,9 @@ static ssize_t devkmsg_read(struct file *file, char __user *buf, logbuf_lock_irq(); } - if (user->seq < prb_first_valid_seq(prb)) { + if (r->info->seq != user->seq) { /* our last seen message is gone, return error and reset */ - user->seq = prb_first_valid_seq(prb); + user->seq = r->info->seq; ret = -EPIPE; logbuf_unlock_irq(); goto out; @@ -812,6 +812,7 @@ static loff_t devkmsg_llseek(struct file *file, loff_t offset, int whence) static __poll_t devkmsg_poll(struct file *file, poll_table *wait) { struct devkmsg_user *user = file->private_data; + struct printk_info info; __poll_t ret = 0; if (!user) @@ -820,9 +821,9 @@ static __poll_t devkmsg_poll(struct file *file, poll_table *wait) poll_wait(file, &log_wait, wait); logbuf_lock_irq(); - if (prb_read_valid(prb, user->seq, NULL)) { + if (prb_read_valid_info(prb, user->seq, &info, NULL)) { /* return error when data has vanished underneath us */ - if (user->seq < prb_first_valid_seq(prb)) + if (info.seq != user->seq) ret = EPOLLIN|EPOLLRDNORM|EPOLLERR|EPOLLPRI; else ret = EPOLLIN|EPOLLRDNORM; @@ -1559,6 +1560,7 @@ static void syslog_clear(void) int do_syslog(int type, char __user *buf, int len, int source) { + struct printk_info info; bool clear = false; static int saved_console_loglevel = LOGLEVEL_DEFAULT; int error; @@ -1629,9 +1631,14 @@ int do_syslog(int type, char __user *buf, int len, int source) /* Number of chars in the log buffer */ case SYSLOG_ACTION_SIZE_UNREAD: logbuf_lock_irq(); - if (syslog_seq < prb_first_valid_seq(prb)) { + if (!prb_read_valid_info(prb, syslog_seq, &info, NULL)) { + /* No unread messages. */ + logbuf_unlock_irq(); + return 0; + } + if (info.seq != syslog_seq) { /* messages are gone, move to first one */ - syslog_seq = prb_first_valid_seq(prb); + syslog_seq = info.seq; syslog_partial = 0; } if (source == SYSLOG_FROM_PROC) { @@ -1643,7 +1650,6 @@ int do_syslog(int type, char __user *buf, int len, int source) error = prb_next_seq(prb) - syslog_seq; } else { bool time = syslog_partial ? syslog_time : printk_time; - struct printk_info info; unsigned int line_count; u64 seq; @@ -3429,9 +3435,11 @@ bool kmsg_dump_get_buffer(struct kmsg_dumper *dumper, bool syslog, goto out; logbuf_lock_irqsave(flags); - if (dumper->cur_seq < prb_first_valid_seq(prb)) { - /* messages are gone, move to first available one */ - dumper->cur_seq = prb_first_valid_seq(prb); + if (prb_read_valid_info(prb, dumper->cur_seq, &info, NULL)) { + if (info.seq != dumper->cur_seq) { + /* messages are gone, move to first available one */ + dumper->cur_seq = info.seq; + } } /* last entry */ diff --git a/kernel/printk/printk_ringbuffer.h b/kernel/printk/printk_ringbuffer.h index 5dc9d022db07..73cc80e01cef 100644 --- a/kernel/printk/printk_ringbuffer.h +++ b/kernel/printk/printk_ringbuffer.h @@ -287,7 +287,7 @@ _DEFINE_PRINTKRB(name, descbits, avgtextbits, &_##name##_text[0]) /* Writer Interface */ /** - * prb_rec_init_wd() - Initialize a buffer for writing records. + * prb_rec_init_wr() - Initialize a buffer for writing records. * * @r: The record to initialize. * @text_buf_size: The needed text buffer size. diff --git a/kernel/printk/printk_safe.c b/kernel/printk/printk_safe.c index a0e6f746de6c..2e9e3ed7d63e 100644 --- a/kernel/printk/printk_safe.c +++ b/kernel/printk/printk_safe.c @@ -45,6 +45,8 @@ struct printk_safe_seq_buf { static DEFINE_PER_CPU(struct printk_safe_seq_buf, safe_print_seq); static DEFINE_PER_CPU(int, printk_context); +static DEFINE_RAW_SPINLOCK(safe_read_lock); + #ifdef CONFIG_PRINTK_NMI static DEFINE_PER_CPU(struct printk_safe_seq_buf, nmi_print_seq); #endif @@ -180,8 +182,6 @@ static void report_message_lost(struct printk_safe_seq_buf *s) */ static void __printk_safe_flush(struct irq_work *work) { - static raw_spinlock_t read_lock = - __RAW_SPIN_LOCK_INITIALIZER(read_lock); struct printk_safe_seq_buf *s = container_of(work, struct printk_safe_seq_buf, work); unsigned long flags; @@ -195,7 +195,7 @@ static void __printk_safe_flush(struct irq_work *work) * different CPUs. This is especially important when printing * a backtrace. */ - raw_spin_lock_irqsave(&read_lock, flags); + raw_spin_lock_irqsave(&safe_read_lock, flags); i = 0; more: @@ -232,7 +232,7 @@ more: out: report_message_lost(s); - raw_spin_unlock_irqrestore(&read_lock, flags); + raw_spin_unlock_irqrestore(&safe_read_lock, flags); } /** @@ -278,6 +278,14 @@ void printk_safe_flush_on_panic(void) raw_spin_lock_init(&logbuf_lock); } + if (raw_spin_is_locked(&safe_read_lock)) { + if (num_online_cpus() > 1) + return; + + debug_locks_off(); + raw_spin_lock_init(&safe_read_lock); + } + printk_safe_flush(); } diff --git a/kernel/ptrace.c b/kernel/ptrace.c index 61db50f7ca86..821cf1723814 100644 --- a/kernel/ptrace.c +++ b/kernel/ptrace.c @@ -375,7 +375,7 @@ static int ptrace_attach(struct task_struct *task, long request, audit_ptrace(task); retval = -EPERM; - if (unlikely(task->flags & PF_KTHREAD)) + if (unlikely(task->flags & (PF_KTHREAD | PF_IO_WORKER))) goto out; if (same_thread_group(task, current)) goto out; diff --git a/kernel/resource.c b/kernel/resource.c index 833394f9c608..627e61b0c124 100644 --- a/kernel/resource.c +++ b/kernel/resource.c @@ -18,12 +18,15 @@ #include <linux/spinlock.h> #include <linux/fs.h> #include <linux/proc_fs.h> +#include <linux/pseudo_fs.h> #include <linux/sched.h> #include <linux/seq_file.h> #include <linux/device.h> #include <linux/pfn.h> #include <linux/mm.h> +#include <linux/mount.h> #include <linux/resource_ext.h> +#include <uapi/linux/magic.h> #include <asm/io.h> @@ -1119,6 +1122,55 @@ resource_size_t resource_alignment(struct resource *res) static DECLARE_WAIT_QUEUE_HEAD(muxed_resource_wait); +static struct inode *iomem_inode; + +#ifdef CONFIG_IO_STRICT_DEVMEM +static void revoke_iomem(struct resource *res) +{ + /* pairs with smp_store_release() in iomem_init_inode() */ + struct inode *inode = smp_load_acquire(&iomem_inode); + + /* + * Check that the initialization has completed. Losing the race + * is ok because it means drivers are claiming resources before + * the fs_initcall level of init and prevent iomem_get_mapping users + * from establishing mappings. + */ + if (!inode) + return; + + /* + * The expectation is that the driver has successfully marked + * the resource busy by this point, so devmem_is_allowed() + * should start returning false, however for performance this + * does not iterate the entire resource range. + */ + if (devmem_is_allowed(PHYS_PFN(res->start)) && + devmem_is_allowed(PHYS_PFN(res->end))) { + /* + * *cringe* iomem=relaxed says "go ahead, what's the + * worst that can happen?" + */ + return; + } + + unmap_mapping_range(inode->i_mapping, res->start, resource_size(res), 1); +} +#else +static void revoke_iomem(struct resource *res) {} +#endif + +struct address_space *iomem_get_mapping(void) +{ + /* + * This function is only called from file open paths, hence guaranteed + * that fs_initcalls have completed and no need to check for NULL. But + * since revoke_iomem can be called before the initcall we still need + * the barrier to appease checkers. + */ + return smp_load_acquire(&iomem_inode)->i_mapping; +} + /** * __request_region - create a new busy resource region * @parent: parent resource descriptor @@ -1186,7 +1238,7 @@ struct resource * __request_region(struct resource *parent, write_unlock(&resource_lock); if (res && orig_parent == &iomem_resource) - revoke_devmem(res); + revoke_iomem(res); return res; } @@ -1786,4 +1838,48 @@ static int __init strict_iomem(char *str) return 1; } +static int iomem_fs_init_fs_context(struct fs_context *fc) +{ + return init_pseudo(fc, DEVMEM_MAGIC) ? 0 : -ENOMEM; +} + +static struct file_system_type iomem_fs_type = { + .name = "iomem", + .owner = THIS_MODULE, + .init_fs_context = iomem_fs_init_fs_context, + .kill_sb = kill_anon_super, +}; + +static int __init iomem_init_inode(void) +{ + static struct vfsmount *iomem_vfs_mount; + static int iomem_fs_cnt; + struct inode *inode; + int rc; + + rc = simple_pin_fs(&iomem_fs_type, &iomem_vfs_mount, &iomem_fs_cnt); + if (rc < 0) { + pr_err("Cannot mount iomem pseudo filesystem: %d\n", rc); + return rc; + } + + inode = alloc_anon_inode(iomem_vfs_mount->mnt_sb); + if (IS_ERR(inode)) { + rc = PTR_ERR(inode); + pr_err("Cannot allocate inode for iomem: %d\n", rc); + simple_release_fs(&iomem_vfs_mount, &iomem_fs_cnt); + return rc; + } + + /* + * Publish iomem revocation inode initialized. + * Pairs with smp_load_acquire() in revoke_iomem(). + */ + smp_store_release(&iomem_inode, inode); + + return 0; +} + +fs_initcall(iomem_init_inode); + __setup("iomem=", strict_iomem); diff --git a/kernel/sched/core.c b/kernel/sched/core.c index 7f5ffc878411..ca2bb629595f 100644 --- a/kernel/sched/core.c +++ b/kernel/sched/core.c @@ -7028,6 +7028,46 @@ int __cond_resched_lock(spinlock_t *lock) } EXPORT_SYMBOL(__cond_resched_lock); +int __cond_resched_rwlock_read(rwlock_t *lock) +{ + int resched = should_resched(PREEMPT_LOCK_OFFSET); + int ret = 0; + + lockdep_assert_held_read(lock); + + if (rwlock_needbreak(lock) || resched) { + read_unlock(lock); + if (resched) + preempt_schedule_common(); + else + cpu_relax(); + ret = 1; + read_lock(lock); + } + return ret; +} +EXPORT_SYMBOL(__cond_resched_rwlock_read); + +int __cond_resched_rwlock_write(rwlock_t *lock) +{ + int resched = should_resched(PREEMPT_LOCK_OFFSET); + int ret = 0; + + lockdep_assert_held_write(lock); + + if (rwlock_needbreak(lock) || resched) { + write_unlock(lock); + if (resched) + preempt_schedule_common(); + else + cpu_relax(); + ret = 1; + write_lock(lock); + } + return ret; +} +EXPORT_SYMBOL(__cond_resched_rwlock_write); + /** * yield - yield the current processor to other threads. * diff --git a/kernel/sched/cpufreq_schedutil.c b/kernel/sched/cpufreq_schedutil.c index 41e498b0008a..50cbad89f7fa 100644 --- a/kernel/sched/cpufreq_schedutil.c +++ b/kernel/sched/cpufreq_schedutil.c @@ -26,7 +26,7 @@ struct sugov_policy { struct sugov_tunables *tunables; struct list_head tunables_hook; - raw_spinlock_t update_lock; /* For shared policies */ + raw_spinlock_t update_lock; u64 last_freq_update_time; s64 freq_update_delay_ns; unsigned int next_freq; @@ -320,23 +320,21 @@ static inline bool sugov_cpu_is_busy(struct sugov_cpu *sg_cpu) { return false; } * Make sugov_should_update_freq() ignore the rate limit when DL * has increased the utilization. */ -static inline void ignore_dl_rate_limit(struct sugov_cpu *sg_cpu, struct sugov_policy *sg_policy) +static inline void ignore_dl_rate_limit(struct sugov_cpu *sg_cpu) { if (cpu_bw_dl(cpu_rq(sg_cpu->cpu)) > sg_cpu->bw_dl) - sg_policy->limits_changed = true; + sg_cpu->sg_policy->limits_changed = true; } static inline bool sugov_update_single_common(struct sugov_cpu *sg_cpu, u64 time, unsigned int flags) { - struct sugov_policy *sg_policy = sg_cpu->sg_policy; - sugov_iowait_boost(sg_cpu, time, flags); sg_cpu->last_update = time; - ignore_dl_rate_limit(sg_cpu, sg_policy); + ignore_dl_rate_limit(sg_cpu); - if (!sugov_should_update_freq(sg_policy, time)) + if (!sugov_should_update_freq(sg_cpu->sg_policy, time)) return false; sugov_get_util(sg_cpu); @@ -451,7 +449,7 @@ sugov_update_shared(struct update_util_data *hook, u64 time, unsigned int flags) sugov_iowait_boost(sg_cpu, time, flags); sg_cpu->last_update = time; - ignore_dl_rate_limit(sg_cpu, sg_policy); + ignore_dl_rate_limit(sg_cpu); if (sugov_should_update_freq(sg_policy, time)) { next_f = sugov_next_freq_shared(sg_cpu, time); diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c index 8a8bd7b13634..794c2cb945f8 100644 --- a/kernel/sched/fair.c +++ b/kernel/sched/fair.c @@ -5126,7 +5126,7 @@ static void do_sched_cfs_slack_timer(struct cfs_bandwidth *cfs_b) /* * When a group wakes up we want to make sure that its quota is not already * expired/exceeded, otherwise it may be allowed to steal additional ticks of - * runtime as update_curr() throttling can not not trigger until it's on-rq. + * runtime as update_curr() throttling can not trigger until it's on-rq. */ static void check_enqueue_throttle(struct cfs_rq *cfs_rq) { diff --git a/kernel/sched/membarrier.c b/kernel/sched/membarrier.c index 08ae45ad9261..acdae625c636 100644 --- a/kernel/sched/membarrier.c +++ b/kernel/sched/membarrier.c @@ -454,7 +454,7 @@ static int sync_runqueues_membarrier_state(struct mm_struct *mm) /* * For each cpu runqueue, if the task's mm match @mm, ensure that all - * @mm's membarrier state set bits are also set in in the runqueue's + * @mm's membarrier state set bits are also set in the runqueue's * membarrier state. This ensures that a runqueue scheduling * between threads which are users of @mm has its membarrier state * updated. diff --git a/kernel/seccomp.c b/kernel/seccomp.c index 952dc1c90229..1d60fc2c9987 100644 --- a/kernel/seccomp.c +++ b/kernel/seccomp.c @@ -1164,7 +1164,7 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd, * Make sure that any changes to mode from another thread have * been seen after SYSCALL_WORK_SECCOMP was seen. */ - rmb(); + smp_rmb(); if (!sd) { populate_seccomp_data(&sd_local); @@ -1284,6 +1284,8 @@ static int __seccomp_filter(int this_syscall, const struct seccomp_data *sd, const bool recheck_after_trace) { BUG(); + + return -1; } #endif diff --git a/kernel/signal.c b/kernel/signal.c index 5ad8566534e7..ba4d1ef39a9e 100644 --- a/kernel/signal.c +++ b/kernel/signal.c @@ -91,7 +91,7 @@ static bool sig_task_ignored(struct task_struct *t, int sig, bool force) return true; /* Only allow kernel generated signals to this kthread */ - if (unlikely((t->flags & PF_KTHREAD) && + if (unlikely((t->flags & (PF_KTHREAD | PF_IO_WORKER)) && (handler == SIG_KTHREAD_KERNEL) && !force)) return true; @@ -1096,7 +1096,7 @@ static int __send_signal(int sig, struct kernel_siginfo *info, struct task_struc /* * Skip useless siginfo allocation for SIGKILL and kernel threads. */ - if ((sig == SIGKILL) || (t->flags & PF_KTHREAD)) + if ((sig == SIGKILL) || (t->flags & (PF_KTHREAD | PF_IO_WORKER))) goto out_set; /* diff --git a/kernel/softirq.c b/kernel/softirq.c index 9d71046ea247..9908ec4a9bfe 100644 --- a/kernel/softirq.c +++ b/kernel/softirq.c @@ -26,6 +26,8 @@ #include <linux/tick.h> #include <linux/irq.h> +#include <asm/softirq_stack.h> + #define CREATE_TRACE_POINTS #include <trace/events/irq.h> diff --git a/kernel/sys.c b/kernel/sys.c index 6928d23c46ea..b09fe21e88ff 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -1242,7 +1242,7 @@ static int override_release(char __user *release, size_t len) break; rest++; } - v = ((LINUX_VERSION_CODE >> 8) & 0xff) + 60; + v = LINUX_VERSION_PATCHLEVEL + 60; copy = clamp_t(size_t, len, 1, sizeof(buf)); copy = scnprintf(buf, copy, "2.6.%u%s", v, rest); ret = copy_to_user(release, buf, copy + 1); @@ -1847,7 +1847,7 @@ static int prctl_set_mm_exe_file(struct mm_struct *mm, unsigned int fd) if (!S_ISREG(inode->i_mode) || path_noexec(&exe.file->f_path)) goto exit; - err = inode_permission(inode, MAY_EXEC); + err = file_permission(exe.file, MAY_EXEC); if (err) goto exit; diff --git a/kernel/sysctl.c b/kernel/sysctl.c index c9fbdd848138..62fbd09b5dc1 100644 --- a/kernel/sysctl.c +++ b/kernel/sysctl.c @@ -2962,7 +2962,7 @@ static struct ctl_table vm_table[] = { .data = &block_dump, .maxlen = sizeof(block_dump), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, { @@ -2970,7 +2970,7 @@ static struct ctl_table vm_table[] = { .data = &sysctl_vfs_cache_pressure, .maxlen = sizeof(sysctl_vfs_cache_pressure), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, #if defined(HAVE_ARCH_PICK_MMAP_LAYOUT) || \ @@ -2980,7 +2980,7 @@ static struct ctl_table vm_table[] = { .data = &sysctl_legacy_va_layout, .maxlen = sizeof(sysctl_legacy_va_layout), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, #endif @@ -2990,7 +2990,7 @@ static struct ctl_table vm_table[] = { .data = &node_reclaim_mode, .maxlen = sizeof(node_reclaim_mode), .mode = 0644, - .proc_handler = proc_dointvec, + .proc_handler = proc_dointvec_minmax, .extra1 = SYSCTL_ZERO, }, { diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig index c1a62ae7e812..7fa82778c3e6 100644 --- a/kernel/trace/Kconfig +++ b/kernel/trace/Kconfig @@ -60,6 +60,11 @@ config HAVE_NOP_MCOUNT help Arch supports the gcc options -pg with -mrecord-mcount and -nop-mcount +config HAVE_OBJTOOL_MCOUNT + bool + help + Arch supports objtool --mcount + config HAVE_C_RECORDMCOUNT bool help @@ -545,7 +550,7 @@ config KPROBE_EVENTS_ON_NOTRACE using kprobe events. If kprobes can use ftrace instead of breakpoint, ftrace related - functions are protected from kprobe-events to prevent an infinit + functions are protected from kprobe-events to prevent an infinite recursion or any unexpected execution path which leads to a kernel crash. @@ -602,6 +607,30 @@ config FTRACE_MCOUNT_RECORD depends on DYNAMIC_FTRACE depends on HAVE_FTRACE_MCOUNT_RECORD +config FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY + bool + depends on FTRACE_MCOUNT_RECORD + +config FTRACE_MCOUNT_USE_CC + def_bool y + depends on $(cc-option,-mrecord-mcount) + depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY + depends on FTRACE_MCOUNT_RECORD + +config FTRACE_MCOUNT_USE_OBJTOOL + def_bool y + depends on HAVE_OBJTOOL_MCOUNT + depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY + depends on !FTRACE_MCOUNT_USE_CC + depends on FTRACE_MCOUNT_RECORD + +config FTRACE_MCOUNT_USE_RECORDMCOUNT + def_bool y + depends on !FTRACE_MCOUNT_USE_PATCHABLE_FUNCTION_ENTRY + depends on !FTRACE_MCOUNT_USE_CC + depends on !FTRACE_MCOUNT_USE_OBJTOOL + depends on FTRACE_MCOUNT_RECORD + config TRACING_MAP bool depends on ARCH_HAVE_NMI_SAFE_CMPXCHG @@ -665,7 +694,7 @@ config TRACEPOINT_BENCHMARK help This option creates the tracepoint "benchmark:benchmark_event". When the tracepoint is enabled, it kicks off a kernel thread that - goes into an infinite loop (calling cond_sched() to let other tasks + goes into an infinite loop (calling cond_resched() to let other tasks run), and calls the tracepoint. Each iteration will record the time it took to write to the tracepoint and the next iteration that data will be passed to the tracepoint itself. That is, the tracepoint @@ -886,6 +915,10 @@ config PREEMPTIRQ_DELAY_TEST irq-disabled critical sections for 500us: modprobe preemptirq_delay_test test_mode=irq delay=500 burst_size=3 + What's more, if you want to attach the test on the cpu which the latency + tracer is running on, specify cpu_affinity=cpu_num at the end of the + command. + If unsure, say N config SYNTH_EVENT_GEN_TEST diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile index 7e44cea89fdc..b28d3e5013cd 100644 --- a/kernel/trace/Makefile +++ b/kernel/trace/Makefile @@ -81,6 +81,7 @@ obj-$(CONFIG_SYNTH_EVENTS) += trace_events_synth.o obj-$(CONFIG_HIST_TRIGGERS) += trace_events_hist.o obj-$(CONFIG_BPF_EVENTS) += bpf_trace.o obj-$(CONFIG_KPROBE_EVENTS) += trace_kprobe.o +obj-$(CONFIG_TRACEPOINTS) += error_report-traces.o obj-$(CONFIG_TRACEPOINTS) += power-traces.o ifeq ($(CONFIG_PM),y) obj-$(CONFIG_TRACEPOINTS) += rpm-traces.o diff --git a/kernel/trace/blktrace.c b/kernel/trace/blktrace.c index 9e9ee4945043..c221e4c3f625 100644 --- a/kernel/trace/blktrace.c +++ b/kernel/trace/blktrace.c @@ -72,17 +72,17 @@ static void trace_note(struct blk_trace *bt, pid_t pid, int action, struct blk_io_trace *t; struct ring_buffer_event *event = NULL; struct trace_buffer *buffer = NULL; - int pc = 0; + unsigned int trace_ctx = 0; int cpu = smp_processor_id(); bool blk_tracer = blk_tracer_enabled; ssize_t cgid_len = cgid ? sizeof(cgid) : 0; if (blk_tracer) { buffer = blk_tr->array_buffer.buffer; - pc = preempt_count(); + trace_ctx = tracing_gen_ctx_flags(0); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, sizeof(*t) + len + cgid_len, - 0, pc); + trace_ctx); if (!event) return; t = ring_buffer_event_data(event); @@ -107,7 +107,7 @@ record_it: memcpy((void *) t + sizeof(*t) + cgid_len, data, len); if (blk_tracer) - trace_buffer_unlock_commit(blk_tr, buffer, event, 0, pc); + trace_buffer_unlock_commit(blk_tr, buffer, event, trace_ctx); } } @@ -222,8 +222,9 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, struct blk_io_trace *t; unsigned long flags = 0; unsigned long *sequence; + unsigned int trace_ctx = 0; pid_t pid; - int cpu, pc = 0; + int cpu; bool blk_tracer = blk_tracer_enabled; ssize_t cgid_len = cgid ? sizeof(cgid) : 0; @@ -252,10 +253,10 @@ static void __blk_add_trace(struct blk_trace *bt, sector_t sector, int bytes, tracing_record_cmdline(current); buffer = blk_tr->array_buffer.buffer; - pc = preempt_count(); + trace_ctx = tracing_gen_ctx_flags(0); event = trace_buffer_lock_reserve(buffer, TRACE_BLK, sizeof(*t) + pdu_len + cgid_len, - 0, pc); + trace_ctx); if (!event) return; t = ring_buffer_event_data(event); @@ -301,7 +302,7 @@ record_it: memcpy((void *)t + sizeof(*t) + cgid_len, pdu_data, pdu_len); if (blk_tracer) { - trace_buffer_unlock_commit(blk_tr, buffer, event, 0, pc); + trace_buffer_unlock_commit(blk_tr, buffer, event, trace_ctx); return; } } @@ -311,8 +312,6 @@ record_it: static void blk_trace_free(struct blk_trace *bt) { - debugfs_remove(bt->msg_file); - debugfs_remove(bt->dropped_file); relay_close(bt->rchan); debugfs_remove(bt->dir); free_percpu(bt->sequence); @@ -544,10 +543,8 @@ static int do_blk_trace_setup(struct request_queue *q, char *name, dev_t dev, INIT_LIST_HEAD(&bt->running_list); ret = -EIO; - bt->dropped_file = debugfs_create_file("dropped", 0444, dir, bt, - &blk_dropped_fops); - - bt->msg_file = debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops); + debugfs_create_file("dropped", 0444, dir, bt, &blk_dropped_fops); + debugfs_create_file("msg", 0222, dir, bt, &blk_msg_fops); bt->rchan = relay_open("trace", dir, buts->buf_size, buts->buf_nr, &blk_relay_callbacks, bt); @@ -1867,7 +1864,17 @@ void blk_trace_remove_sysfs(struct device *dev) #ifdef CONFIG_EVENT_TRACING -void blk_fill_rwbs(char *rwbs, unsigned int op, int bytes) +/** + * blk_fill_rwbs - Fill the buffer rwbs by mapping op to character string. + * @rwbs: buffer to be filled + * @op: REQ_OP_XXX for the tracepoint + * + * Description: + * Maps the REQ_OP_XXX to character and fills the buffer provided by the + * caller with resulting string. + * + **/ +void blk_fill_rwbs(char *rwbs, unsigned int op) { int i = 0; diff --git a/kernel/trace/error_report-traces.c b/kernel/trace/error_report-traces.c new file mode 100644 index 000000000000..f89792c25b11 --- /dev/null +++ b/kernel/trace/error_report-traces.c @@ -0,0 +1,11 @@ +// SPDX-License-Identifier: GPL-2.0 +/* + * Error reporting trace points. + * + * Copyright (C) 2021, Google LLC. + */ + +#define CREATE_TRACE_POINTS +#include <trace/events/error_report.h> + +EXPORT_TRACEPOINT_SYMBOL_GPL(error_report_end); diff --git a/kernel/trace/preemptirq_delay_test.c b/kernel/trace/preemptirq_delay_test.c index 312d1a0ca3b6..8c4ffd076162 100644 --- a/kernel/trace/preemptirq_delay_test.c +++ b/kernel/trace/preemptirq_delay_test.c @@ -21,13 +21,16 @@ static ulong delay = 100; static char test_mode[12] = "irq"; static uint burst_size = 1; +static int cpu_affinity = -1; module_param_named(delay, delay, ulong, 0444); module_param_string(test_mode, test_mode, 12, 0444); module_param_named(burst_size, burst_size, uint, 0444); +module_param_named(cpu_affinity, cpu_affinity, int, 0444); MODULE_PARM_DESC(delay, "Period in microseconds (100 us default)"); MODULE_PARM_DESC(test_mode, "Mode of the test such as preempt, irq, or alternate (default irq)"); MODULE_PARM_DESC(burst_size, "The size of a burst (default 1)"); +MODULE_PARM_DESC(cpu_affinity, "Cpu num test is running on"); static struct completion done; @@ -36,7 +39,9 @@ static struct completion done; static void busy_wait(ulong time) { u64 start, end; + start = trace_clock_local(); + do { end = trace_clock_local(); if (kthread_should_stop()) @@ -47,6 +52,7 @@ static void busy_wait(ulong time) static __always_inline void irqoff_test(void) { unsigned long flags; + local_irq_save(flags); busy_wait(delay); local_irq_restore(flags); @@ -113,6 +119,14 @@ static int preemptirq_delay_run(void *data) { int i; int s = MIN(burst_size, NR_TEST_FUNCS); + struct cpumask cpu_mask; + + if (cpu_affinity > -1) { + cpumask_clear(&cpu_mask); + cpumask_set_cpu(cpu_affinity, &cpu_mask); + if (set_cpus_allowed_ptr(current, &cpu_mask)) + pr_err("cpu_affinity:%d, failed\n", cpu_affinity); + } for (i = 0; i < s; i++) (testfuncs[i])(i); diff --git a/kernel/trace/ring_buffer.c b/kernel/trace/ring_buffer.c index ec08f948dd80..68744c51517e 100644 --- a/kernel/trace/ring_buffer.c +++ b/kernel/trace/ring_buffer.c @@ -1112,8 +1112,7 @@ static struct list_head *rb_list_head(struct list_head *list) * its flags will be non zero. */ static inline int -rb_is_head_page(struct ring_buffer_per_cpu *cpu_buffer, - struct buffer_page *page, struct list_head *list) +rb_is_head_page(struct buffer_page *page, struct list_head *list) { unsigned long val; @@ -1142,8 +1141,7 @@ static bool rb_is_reader_page(struct buffer_page *page) /* * rb_set_list_to_head - set a list_head to be pointing to head. */ -static void rb_set_list_to_head(struct ring_buffer_per_cpu *cpu_buffer, - struct list_head *list) +static void rb_set_list_to_head(struct list_head *list) { unsigned long *ptr; @@ -1166,7 +1164,7 @@ static void rb_head_page_activate(struct ring_buffer_per_cpu *cpu_buffer) /* * Set the previous list pointer to have the HEAD flag. */ - rb_set_list_to_head(cpu_buffer, head->list.prev); + rb_set_list_to_head(head->list.prev); } static void rb_list_head_clear(struct list_head *list) @@ -1241,8 +1239,7 @@ static int rb_head_page_set_normal(struct ring_buffer_per_cpu *cpu_buffer, old_flag, RB_PAGE_NORMAL); } -static inline void rb_inc_page(struct ring_buffer_per_cpu *cpu_buffer, - struct buffer_page **bpage) +static inline void rb_inc_page(struct buffer_page **bpage) { struct list_head *p = rb_list_head((*bpage)->list.next); @@ -1274,11 +1271,11 @@ rb_set_head_page(struct ring_buffer_per_cpu *cpu_buffer) */ for (i = 0; i < 3; i++) { do { - if (rb_is_head_page(cpu_buffer, page, page->list.prev)) { + if (rb_is_head_page(page, page->list.prev)) { cpu_buffer->head_page = page; return page; } - rb_inc_page(cpu_buffer, &page); + rb_inc_page(&page); } while (page != head); } @@ -1824,7 +1821,7 @@ rb_remove_pages(struct ring_buffer_per_cpu *cpu_buffer, unsigned long nr_pages) cond_resched(); to_remove_page = tmp_iter_page; - rb_inc_page(cpu_buffer, &tmp_iter_page); + rb_inc_page(&tmp_iter_page); /* update the counters */ page_entries = rb_page_entries(to_remove_page); @@ -2062,10 +2059,6 @@ int ring_buffer_resize(struct trace_buffer *buffer, unsigned long size, put_online_cpus(); } else { - /* Make sure this CPU has been initialized */ - if (!cpumask_test_cpu(cpu_id, buffer->cpumask)) - goto out; - cpu_buffer = buffer->buffers[cpu_id]; if (nr_pages == cpu_buffer->nr_pages) @@ -2271,7 +2264,7 @@ static void rb_inc_iter(struct ring_buffer_iter *iter) if (iter->head_page == cpu_buffer->reader_page) iter->head_page = rb_set_head_page(cpu_buffer); else - rb_inc_page(cpu_buffer, &iter->head_page); + rb_inc_page(&iter->head_page); iter->page_stamp = iter->read_stamp = iter->head_page->page->time_stamp; iter->head = 0; @@ -2374,7 +2367,7 @@ rb_handle_head_page(struct ring_buffer_per_cpu *cpu_buffer, * want the outer most commit to reset it. */ new_head = next_page; - rb_inc_page(cpu_buffer, &new_head); + rb_inc_page(&new_head); ret = rb_head_page_set_head(cpu_buffer, new_head, next_page, RB_PAGE_NORMAL); @@ -2526,7 +2519,7 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, next_page = tail_page; - rb_inc_page(cpu_buffer, &next_page); + rb_inc_page(&next_page); /* * If for some reason, we had an interrupt storm that made @@ -2552,7 +2545,7 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, * the buffer, unless the commit page is still on the * reader page. */ - if (rb_is_head_page(cpu_buffer, next_page, &tail_page->list)) { + if (rb_is_head_page(next_page, &tail_page->list)) { /* * If the commit is not on the reader page, then @@ -2583,7 +2576,7 @@ rb_move_tail(struct ring_buffer_per_cpu *cpu_buffer, * have filled up the buffer with events * from interrupts and such, and wrapped. * - * Note, if the tail page is also the on the + * Note, if the tail page is also on the * reader_page, we let it move out. */ if (unlikely((cpu_buffer->commit_page != @@ -2822,6 +2815,17 @@ rb_try_to_discard(struct ring_buffer_per_cpu *cpu_buffer, return 0; /* + * It's possible that the event time delta is zero + * (has the same time stamp as the previous event) + * in which case write_stamp and before_stamp could + * be the same. In such a case, force before_stamp + * to be different than write_stamp. It doesn't + * matter what it is, as long as its different. + */ + if (!delta) + rb_time_set(&cpu_buffer->before_stamp, 0); + + /* * If an event were to come in now, it would see that the * write_stamp and the before_stamp are different, and assume * that this event just added itself before updating @@ -2879,7 +2883,7 @@ rb_set_commit_to_write(struct ring_buffer_per_cpu *cpu_buffer) return; local_set(&cpu_buffer->commit_page->page->commit, rb_page_write(cpu_buffer->commit_page)); - rb_inc_page(cpu_buffer, &cpu_buffer->commit_page); + rb_inc_page(&cpu_buffer->commit_page); /* add barrier to keep gcc from optimizing too much */ barrier(); } @@ -3314,9 +3318,13 @@ static void check_buffer(struct ring_buffer_per_cpu *cpu_buffer, goto out; } atomic_inc(&cpu_buffer->record_disabled); - pr_warn("[CPU: %d]TIME DOES NOT MATCH expected:%lld actual:%lld delta:%lld after:%lld\n", - cpu_buffer->cpu, - ts + info->delta, info->ts, info->delta, info->after); + /* There's some cases in boot up that this can happen */ + WARN_ON_ONCE(system_state != SYSTEM_BOOTING); + pr_warn("[CPU: %d]TIME DOES NOT MATCH expected:%lld actual:%lld delta:%lld before:%lld after:%lld%s\n", + cpu_buffer->cpu, + ts + info->delta, info->ts, info->delta, + info->before, info->after, + full ? " (full)" : ""); dump_buffer_page(bpage, info, tail); atomic_dec(&ts_dump); /* Do not re-enable checking */ @@ -3638,14 +3646,14 @@ rb_decrement_entry(struct ring_buffer_per_cpu *cpu_buffer, * Because the commit page may be on the reader page we * start with the next page and check the end loop there. */ - rb_inc_page(cpu_buffer, &bpage); + rb_inc_page(&bpage); start = bpage; do { if (bpage->page == (void *)addr) { local_dec(&bpage->entries); return; } - rb_inc_page(cpu_buffer, &bpage); + rb_inc_page(&bpage); } while (bpage != start); /* commit not part of this buffer?? */ @@ -4367,7 +4375,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) cpu_buffer->pages = reader->list.prev; /* The reader page will be pointing to the new head */ - rb_set_list_to_head(cpu_buffer, &cpu_buffer->reader_page->list); + rb_set_list_to_head(&cpu_buffer->reader_page->list); /* * We want to make sure we read the overruns after we set up our @@ -4406,7 +4414,7 @@ rb_get_reader_page(struct ring_buffer_per_cpu *cpu_buffer) * Now make the new head point back to the reader page. */ rb_list_head(reader->list.next)->prev = &cpu_buffer->reader_page->list; - rb_inc_page(cpu_buffer, &cpu_buffer->head_page); + rb_inc_page(&cpu_buffer->head_page); local_inc(&cpu_buffer->pages_read); diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c index b5815a022ecc..eccb4e1187cc 100644 --- a/kernel/trace/trace.c +++ b/kernel/trace/trace.c @@ -176,7 +176,7 @@ static union trace_eval_map_item *trace_eval_maps; int tracing_set_tracer(struct trace_array *tr, const char *buf); static void ftrace_trace_userstack(struct trace_array *tr, struct trace_buffer *buffer, - unsigned long flags, int pc); + unsigned int trace_ctx); #define MAX_TRACER_SIZE 100 static char bootup_tracer_buf[MAX_TRACER_SIZE] __initdata; @@ -408,7 +408,8 @@ EXPORT_SYMBOL_GPL(unregister_ftrace_export); TRACE_ITER_PRINT_PARENT | TRACE_ITER_PRINTK | \ TRACE_ITER_ANNOTATE | TRACE_ITER_CONTEXT_INFO | \ TRACE_ITER_RECORD_CMD | TRACE_ITER_OVERWRITE | \ - TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS) + TRACE_ITER_IRQ_INFO | TRACE_ITER_MARKERS | \ + TRACE_ITER_HASH_PTR) /* trace_options that are only supported by global_trace */ #define TOP_LEVEL_TRACE_FLAGS (TRACE_ITER_PRINTK | \ @@ -454,6 +455,7 @@ static void __trace_array_put(struct trace_array *this_tr) /** * trace_array_put - Decrement the reference counter for this trace array. + * @this_tr : pointer to the trace array * * NOTE: Use this when we no longer need the trace array returned by * trace_array_get_by_name(). This ensures the trace array can be later @@ -530,6 +532,7 @@ trace_find_filtered_pid(struct trace_pid_list *filtered_pids, pid_t search_pid) /** * trace_ignore_this_task - should a task be ignored for tracing * @filtered_pids: The list of pids to check + * @filtered_no_pids: The list of pids not to be traced * @task: The task that should be ignored if not filtered * * Checks if @task should be traced or not from @filtered_pids. @@ -780,7 +783,7 @@ u64 ftrace_now(int cpu) } /** - * tracing_is_enabled - Show if global_trace has been disabled + * tracing_is_enabled - Show if global_trace has been enabled * * Shows if the global trace has been enabled or not. It uses the * mirror flag "buffer_disabled" to be used in fast paths such as for @@ -905,23 +908,23 @@ static inline void trace_access_lock_init(void) #ifdef CONFIG_STACKTRACE static void __ftrace_trace_stack(struct trace_buffer *buffer, - unsigned long flags, - int skip, int pc, struct pt_regs *regs); + unsigned int trace_ctx, + int skip, struct pt_regs *regs); static inline void ftrace_trace_stack(struct trace_array *tr, struct trace_buffer *buffer, - unsigned long flags, - int skip, int pc, struct pt_regs *regs); + unsigned int trace_ctx, + int skip, struct pt_regs *regs); #else static inline void __ftrace_trace_stack(struct trace_buffer *buffer, - unsigned long flags, - int skip, int pc, struct pt_regs *regs) + unsigned int trace_ctx, + int skip, struct pt_regs *regs) { } static inline void ftrace_trace_stack(struct trace_array *tr, struct trace_buffer *buffer, - unsigned long flags, - int skip, int pc, struct pt_regs *regs) + unsigned long trace_ctx, + int skip, struct pt_regs *regs) { } @@ -929,24 +932,24 @@ static inline void ftrace_trace_stack(struct trace_array *tr, static __always_inline void trace_event_setup(struct ring_buffer_event *event, - int type, unsigned long flags, int pc) + int type, unsigned int trace_ctx) { struct trace_entry *ent = ring_buffer_event_data(event); - tracing_generic_entry_update(ent, type, flags, pc); + tracing_generic_entry_update(ent, type, trace_ctx); } static __always_inline struct ring_buffer_event * __trace_buffer_lock_reserve(struct trace_buffer *buffer, int type, unsigned long len, - unsigned long flags, int pc) + unsigned int trace_ctx) { struct ring_buffer_event *event; event = ring_buffer_lock_reserve(buffer, len); if (event != NULL) - trace_event_setup(event, type, flags, pc); + trace_event_setup(event, type, trace_ctx); return event; } @@ -1007,25 +1010,22 @@ int __trace_puts(unsigned long ip, const char *str, int size) struct ring_buffer_event *event; struct trace_buffer *buffer; struct print_entry *entry; - unsigned long irq_flags; + unsigned int trace_ctx; int alloc; - int pc; if (!(global_trace.trace_flags & TRACE_ITER_PRINTK)) return 0; - pc = preempt_count(); - if (unlikely(tracing_selftest_running || tracing_disabled)) return 0; alloc = sizeof(*entry) + size + 2; /* possible \n added */ - local_save_flags(irq_flags); + trace_ctx = tracing_gen_ctx(); buffer = global_trace.array_buffer.buffer; ring_buffer_nest_start(buffer); - event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, alloc, - irq_flags, pc); + event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, alloc, + trace_ctx); if (!event) { size = 0; goto out; @@ -1044,7 +1044,7 @@ int __trace_puts(unsigned long ip, const char *str, int size) entry->buf[size] = '\0'; __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(&global_trace, buffer, irq_flags, 4, pc, NULL); + ftrace_trace_stack(&global_trace, buffer, trace_ctx, 4, NULL); out: ring_buffer_nest_end(buffer); return size; @@ -1061,25 +1061,22 @@ int __trace_bputs(unsigned long ip, const char *str) struct ring_buffer_event *event; struct trace_buffer *buffer; struct bputs_entry *entry; - unsigned long irq_flags; + unsigned int trace_ctx; int size = sizeof(struct bputs_entry); int ret = 0; - int pc; if (!(global_trace.trace_flags & TRACE_ITER_PRINTK)) return 0; - pc = preempt_count(); - if (unlikely(tracing_selftest_running || tracing_disabled)) return 0; - local_save_flags(irq_flags); + trace_ctx = tracing_gen_ctx(); buffer = global_trace.array_buffer.buffer; ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_BPUTS, size, - irq_flags, pc); + trace_ctx); if (!event) goto out; @@ -1088,7 +1085,7 @@ int __trace_bputs(unsigned long ip, const char *str) entry->str = str; __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(&global_trace, buffer, irq_flags, 4, pc, NULL); + ftrace_trace_stack(&global_trace, buffer, trace_ctx, 4, NULL); ret = 1; out: @@ -1932,6 +1929,12 @@ static int run_tracer_selftest(struct tracer *type) if (!selftests_can_run) return save_selftest(type); + if (!tracing_is_on()) { + pr_warn("Selftest for tracer %s skipped due to tracing disabled\n", + type->name); + return 0; + } + /* * Run a selftest on this tracer. * Here we reset the trace buffer, and set the current @@ -2584,36 +2587,34 @@ enum print_line_t trace_handle_return(struct trace_seq *s) } EXPORT_SYMBOL_GPL(trace_handle_return); -void -tracing_generic_entry_update(struct trace_entry *entry, unsigned short type, - unsigned long flags, int pc) +unsigned int tracing_gen_ctx_irq_test(unsigned int irqs_status) { - struct task_struct *tsk = current; + unsigned int trace_flags = irqs_status; + unsigned int pc; - entry->preempt_count = pc & 0xff; - entry->pid = (tsk) ? tsk->pid : 0; - entry->type = type; - entry->flags = -#ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT - (irqs_disabled_flags(flags) ? TRACE_FLAG_IRQS_OFF : 0) | -#else - TRACE_FLAG_IRQS_NOSUPPORT | -#endif - ((pc & NMI_MASK ) ? TRACE_FLAG_NMI : 0) | - ((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) | - ((pc & SOFTIRQ_OFFSET) ? TRACE_FLAG_SOFTIRQ : 0) | - (tif_need_resched() ? TRACE_FLAG_NEED_RESCHED : 0) | - (test_preempt_need_resched() ? TRACE_FLAG_PREEMPT_RESCHED : 0); + pc = preempt_count(); + + if (pc & NMI_MASK) + trace_flags |= TRACE_FLAG_NMI; + if (pc & HARDIRQ_MASK) + trace_flags |= TRACE_FLAG_HARDIRQ; + if (in_serving_softirq()) + trace_flags |= TRACE_FLAG_SOFTIRQ; + + if (tif_need_resched()) + trace_flags |= TRACE_FLAG_NEED_RESCHED; + if (test_preempt_need_resched()) + trace_flags |= TRACE_FLAG_PREEMPT_RESCHED; + return (trace_flags << 16) | (pc & 0xff); } -EXPORT_SYMBOL_GPL(tracing_generic_entry_update); struct ring_buffer_event * trace_buffer_lock_reserve(struct trace_buffer *buffer, int type, unsigned long len, - unsigned long flags, int pc) + unsigned int trace_ctx) { - return __trace_buffer_lock_reserve(buffer, type, len, flags, pc); + return __trace_buffer_lock_reserve(buffer, type, len, trace_ctx); } DEFINE_PER_CPU(struct ring_buffer_event *, trace_buffered_event); @@ -2733,7 +2734,7 @@ struct ring_buffer_event * trace_event_buffer_lock_reserve(struct trace_buffer **current_rb, struct trace_event_file *trace_file, int type, unsigned long len, - unsigned long flags, int pc) + unsigned int trace_ctx) { struct ring_buffer_event *entry; int val; @@ -2746,15 +2747,15 @@ trace_event_buffer_lock_reserve(struct trace_buffer **current_rb, /* Try to use the per cpu buffer first */ val = this_cpu_inc_return(trace_buffered_event_cnt); if ((len < (PAGE_SIZE - sizeof(*entry))) && val == 1) { - trace_event_setup(entry, type, flags, pc); + trace_event_setup(entry, type, trace_ctx); entry->array[0] = len; return entry; } this_cpu_dec(trace_buffered_event_cnt); } - entry = __trace_buffer_lock_reserve(*current_rb, - type, len, flags, pc); + entry = __trace_buffer_lock_reserve(*current_rb, type, len, + trace_ctx); /* * If tracing is off, but we have triggers enabled * we still need to look at the event data. Use the temp_buffer @@ -2763,8 +2764,8 @@ trace_event_buffer_lock_reserve(struct trace_buffer **current_rb, */ if (!entry && trace_file->flags & EVENT_FILE_FL_TRIGGER_COND) { *current_rb = temp_buffer; - entry = __trace_buffer_lock_reserve(*current_rb, - type, len, flags, pc); + entry = __trace_buffer_lock_reserve(*current_rb, type, len, + trace_ctx); } return entry; } @@ -2850,7 +2851,7 @@ void trace_event_buffer_commit(struct trace_event_buffer *fbuffer) ftrace_exports(fbuffer->event, TRACE_EXPORT_EVENT); event_trigger_unlock_commit_regs(fbuffer->trace_file, fbuffer->buffer, fbuffer->event, fbuffer->entry, - fbuffer->flags, fbuffer->pc, fbuffer->regs); + fbuffer->trace_ctx, fbuffer->regs); } EXPORT_SYMBOL_GPL(trace_event_buffer_commit); @@ -2866,7 +2867,7 @@ EXPORT_SYMBOL_GPL(trace_event_buffer_commit); void trace_buffer_unlock_commit_regs(struct trace_array *tr, struct trace_buffer *buffer, struct ring_buffer_event *event, - unsigned long flags, int pc, + unsigned int trace_ctx, struct pt_regs *regs) { __buffer_unlock_commit(buffer, event); @@ -2877,8 +2878,8 @@ void trace_buffer_unlock_commit_regs(struct trace_array *tr, * and mmiotrace, but that's ok if they lose a function or * two. They are not that meaningful. */ - ftrace_trace_stack(tr, buffer, flags, regs ? 0 : STACK_SKIP, pc, regs); - ftrace_trace_userstack(tr, buffer, flags, pc); + ftrace_trace_stack(tr, buffer, trace_ctx, regs ? 0 : STACK_SKIP, regs); + ftrace_trace_userstack(tr, buffer, trace_ctx); } /* @@ -2892,9 +2893,8 @@ trace_buffer_unlock_commit_nostack(struct trace_buffer *buffer, } void -trace_function(struct trace_array *tr, - unsigned long ip, unsigned long parent_ip, unsigned long flags, - int pc) +trace_function(struct trace_array *tr, unsigned long ip, unsigned long + parent_ip, unsigned int trace_ctx) { struct trace_event_call *call = &event_function; struct trace_buffer *buffer = tr->array_buffer.buffer; @@ -2902,7 +2902,7 @@ trace_function(struct trace_array *tr, struct ftrace_entry *entry; event = __trace_buffer_lock_reserve(buffer, TRACE_FN, sizeof(*entry), - flags, pc); + trace_ctx); if (!event) return; entry = ring_buffer_event_data(event); @@ -2936,8 +2936,8 @@ static DEFINE_PER_CPU(struct ftrace_stacks, ftrace_stacks); static DEFINE_PER_CPU(int, ftrace_stack_reserve); static void __ftrace_trace_stack(struct trace_buffer *buffer, - unsigned long flags, - int skip, int pc, struct pt_regs *regs) + unsigned int trace_ctx, + int skip, struct pt_regs *regs) { struct trace_event_call *call = &event_kernel_stack; struct ring_buffer_event *event; @@ -2984,7 +2984,7 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, size = nr_entries * sizeof(unsigned long); event = __trace_buffer_lock_reserve(buffer, TRACE_STACK, - sizeof(*entry) + size, flags, pc); + sizeof(*entry) + size, trace_ctx); if (!event) goto out; entry = ring_buffer_event_data(event); @@ -3005,22 +3005,22 @@ static void __ftrace_trace_stack(struct trace_buffer *buffer, static inline void ftrace_trace_stack(struct trace_array *tr, struct trace_buffer *buffer, - unsigned long flags, - int skip, int pc, struct pt_regs *regs) + unsigned int trace_ctx, + int skip, struct pt_regs *regs) { if (!(tr->trace_flags & TRACE_ITER_STACKTRACE)) return; - __ftrace_trace_stack(buffer, flags, skip, pc, regs); + __ftrace_trace_stack(buffer, trace_ctx, skip, regs); } -void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, - int pc) +void __trace_stack(struct trace_array *tr, unsigned int trace_ctx, + int skip) { struct trace_buffer *buffer = tr->array_buffer.buffer; if (rcu_is_watching()) { - __ftrace_trace_stack(buffer, flags, skip, pc, NULL); + __ftrace_trace_stack(buffer, trace_ctx, skip, NULL); return; } @@ -3034,7 +3034,7 @@ void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, return; rcu_irq_enter_irqson(); - __ftrace_trace_stack(buffer, flags, skip, pc, NULL); + __ftrace_trace_stack(buffer, trace_ctx, skip, NULL); rcu_irq_exit_irqson(); } @@ -3044,19 +3044,15 @@ void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, */ void trace_dump_stack(int skip) { - unsigned long flags; - if (tracing_disabled || tracing_selftest_running) return; - local_save_flags(flags); - #ifndef CONFIG_UNWINDER_ORC /* Skip 1 to skip this function. */ skip++; #endif __ftrace_trace_stack(global_trace.array_buffer.buffer, - flags, skip, preempt_count(), NULL); + tracing_gen_ctx(), skip, NULL); } EXPORT_SYMBOL_GPL(trace_dump_stack); @@ -3065,7 +3061,7 @@ static DEFINE_PER_CPU(int, user_stack_count); static void ftrace_trace_userstack(struct trace_array *tr, - struct trace_buffer *buffer, unsigned long flags, int pc) + struct trace_buffer *buffer, unsigned int trace_ctx) { struct trace_event_call *call = &event_user_stack; struct ring_buffer_event *event; @@ -3092,7 +3088,7 @@ ftrace_trace_userstack(struct trace_array *tr, __this_cpu_inc(user_stack_count); event = __trace_buffer_lock_reserve(buffer, TRACE_USER_STACK, - sizeof(*entry), flags, pc); + sizeof(*entry), trace_ctx); if (!event) goto out_drop_count; entry = ring_buffer_event_data(event); @@ -3112,7 +3108,7 @@ ftrace_trace_userstack(struct trace_array *tr, #else /* CONFIG_USER_STACKTRACE_SUPPORT */ static void ftrace_trace_userstack(struct trace_array *tr, struct trace_buffer *buffer, - unsigned long flags, int pc) + unsigned int trace_ctx) { } #endif /* !CONFIG_USER_STACKTRACE_SUPPORT */ @@ -3242,9 +3238,9 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) struct trace_buffer *buffer; struct trace_array *tr = &global_trace; struct bprint_entry *entry; - unsigned long flags; + unsigned int trace_ctx; char *tbuffer; - int len = 0, size, pc; + int len = 0, size; if (unlikely(tracing_selftest_running || tracing_disabled)) return 0; @@ -3252,7 +3248,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) /* Don't pollute graph traces with trace_vprintk internals */ pause_graph_tracing(); - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); preempt_disable_notrace(); tbuffer = get_trace_buf(); @@ -3266,12 +3262,11 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) if (len > TRACE_BUF_SIZE/sizeof(int) || len < 0) goto out_put; - local_save_flags(flags); size = sizeof(*entry) + sizeof(u32) * len; buffer = tr->array_buffer.buffer; ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_BPRINT, size, - flags, pc); + trace_ctx); if (!event) goto out; entry = ring_buffer_event_data(event); @@ -3281,7 +3276,7 @@ int trace_vbprintk(unsigned long ip, const char *fmt, va_list args) memcpy(entry->buf, tbuffer, sizeof(u32) * len); if (!call_filter_check_discard(call, entry, buffer, event)) { __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(tr, buffer, flags, 6, pc, NULL); + ftrace_trace_stack(tr, buffer, trace_ctx, 6, NULL); } out: @@ -3304,9 +3299,9 @@ __trace_array_vprintk(struct trace_buffer *buffer, { struct trace_event_call *call = &event_print; struct ring_buffer_event *event; - int len = 0, size, pc; + int len = 0, size; struct print_entry *entry; - unsigned long flags; + unsigned int trace_ctx; char *tbuffer; if (tracing_disabled || tracing_selftest_running) @@ -3315,7 +3310,7 @@ __trace_array_vprintk(struct trace_buffer *buffer, /* Don't pollute graph traces with trace_vprintk internals */ pause_graph_tracing(); - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); preempt_disable_notrace(); @@ -3327,11 +3322,10 @@ __trace_array_vprintk(struct trace_buffer *buffer, len = vscnprintf(tbuffer, TRACE_BUF_SIZE, fmt, args); - local_save_flags(flags); size = sizeof(*entry) + len + 1; ring_buffer_nest_start(buffer); event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, - flags, pc); + trace_ctx); if (!event) goto out; entry = ring_buffer_event_data(event); @@ -3340,7 +3334,7 @@ __trace_array_vprintk(struct trace_buffer *buffer, memcpy(&entry->buf, tbuffer, len + 1); if (!call_filter_check_discard(call, entry, buffer, event)) { __buffer_unlock_commit(buffer, event); - ftrace_trace_stack(&global_trace, buffer, flags, 6, pc, NULL); + ftrace_trace_stack(&global_trace, buffer, trace_ctx, 6, NULL); } out: @@ -3543,6 +3537,65 @@ __find_next_entry(struct trace_iterator *iter, int *ent_cpu, return next; } +#define STATIC_FMT_BUF_SIZE 128 +static char static_fmt_buf[STATIC_FMT_BUF_SIZE]; + +static char *trace_iter_expand_format(struct trace_iterator *iter) +{ + char *tmp; + + if (iter->fmt == static_fmt_buf) + return NULL; + + tmp = krealloc(iter->fmt, iter->fmt_size + STATIC_FMT_BUF_SIZE, + GFP_KERNEL); + if (tmp) { + iter->fmt_size += STATIC_FMT_BUF_SIZE; + iter->fmt = tmp; + } + + return tmp; +} + +const char *trace_event_format(struct trace_iterator *iter, const char *fmt) +{ + const char *p, *new_fmt; + char *q; + + if (WARN_ON_ONCE(!fmt)) + return fmt; + + if (iter->tr->trace_flags & TRACE_ITER_HASH_PTR) + return fmt; + + p = fmt; + new_fmt = q = iter->fmt; + while (*p) { + if (unlikely(q - new_fmt + 3 > iter->fmt_size)) { + if (!trace_iter_expand_format(iter)) + return fmt; + + q += iter->fmt - new_fmt; + new_fmt = iter->fmt; + } + + *q++ = *p++; + + /* Replace %p with %px */ + if (p[-1] == '%') { + if (p[0] == '%') { + *q++ = *p++; + } else if (p[0] == 'p' && !isalnum(p[1])) { + *q++ = *p++; + *q++ = 'x'; + } + } + } + *q = '\0'; + + return new_fmt; +} + #define STATIC_TEMP_BUF_SIZE 128 static char static_temp_buf[STATIC_TEMP_BUF_SIZE] __aligned(4); @@ -4336,6 +4389,16 @@ __tracing_open(struct inode *inode, struct file *file, bool snapshot) iter->temp_size = 128; /* + * trace_event_printf() may need to modify given format + * string to replace %p with %px so that it shows real address + * instead of hash value. However, that is only for the event + * tracing, other tracer may not need. Defer the allocation + * until it is needed. + */ + iter->fmt = NULL; + iter->fmt_size = 0; + + /* * We make a copy of the current tracer to avoid concurrent * changes on it while we are reading. */ @@ -4486,6 +4549,7 @@ static int tracing_release(struct inode *inode, struct file *file) mutex_destroy(&iter->mutex); free_cpumask_var(iter->started); + kfree(iter->fmt); kfree(iter->temp); kfree(iter->trace); kfree(iter->buffer_iter); @@ -6653,7 +6717,6 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, enum event_trigger_type tt = ETT_NONE; struct trace_buffer *buffer; struct print_entry *entry; - unsigned long irq_flags; ssize_t written; int size; int len; @@ -6673,7 +6736,6 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, BUILD_BUG_ON(TRACE_BUF_SIZE >= PAGE_SIZE); - local_save_flags(irq_flags); size = sizeof(*entry) + cnt + 2; /* add '\0' and possible '\n' */ /* If less than "<faulted>", then make sure we can still add that */ @@ -6682,7 +6744,7 @@ tracing_mark_write(struct file *filp, const char __user *ubuf, buffer = tr->array_buffer.buffer; event = __trace_buffer_lock_reserve(buffer, TRACE_PRINT, size, - irq_flags, preempt_count()); + tracing_gen_ctx()); if (unlikely(!event)) /* Ring buffer disabled, return as if not open for write */ return -EBADF; @@ -6734,7 +6796,6 @@ tracing_mark_raw_write(struct file *filp, const char __user *ubuf, struct ring_buffer_event *event; struct trace_buffer *buffer; struct raw_data_entry *entry; - unsigned long irq_flags; ssize_t written; int size; int len; @@ -6756,14 +6817,13 @@ tracing_mark_raw_write(struct file *filp, const char __user *ubuf, BUILD_BUG_ON(TRACE_BUF_SIZE >= PAGE_SIZE); - local_save_flags(irq_flags); size = sizeof(*entry) + cnt; if (cnt < FAULT_SIZE_ID) size += FAULT_SIZE_ID - cnt; buffer = tr->array_buffer.buffer; event = __trace_buffer_lock_reserve(buffer, TRACE_RAW_DATA, size, - irq_flags, preempt_count()); + tracing_gen_ctx()); if (!event) /* Ring buffer disabled, return as if not open for write */ return -EBADF; @@ -9348,9 +9408,11 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) /* Simulate the iterator */ trace_init_global_iter(&iter); - /* Can not use kmalloc for iter.temp */ + /* Can not use kmalloc for iter.temp and iter.fmt */ iter.temp = static_temp_buf; iter.temp_size = STATIC_TEMP_BUF_SIZE; + iter.fmt = static_fmt_buf; + iter.fmt_size = STATIC_FMT_BUF_SIZE; for_each_tracing_cpu(cpu) { atomic_inc(&per_cpu_ptr(iter.array_buffer->data, cpu)->disabled); @@ -9429,30 +9491,11 @@ void ftrace_dump(enum ftrace_dump_mode oops_dump_mode) } EXPORT_SYMBOL_GPL(ftrace_dump); -int trace_run_command(const char *buf, int (*createfn)(int, char **)) -{ - char **argv; - int argc, ret; - - argc = 0; - ret = 0; - argv = argv_split(GFP_KERNEL, buf, &argc); - if (!argv) - return -ENOMEM; - - if (argc) - ret = createfn(argc, argv); - - argv_free(argv); - - return ret; -} - #define WRITE_BUFSIZE 4096 ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, size_t count, loff_t *ppos, - int (*createfn)(int, char **)) + int (*createfn)(const char *)) { char *kbuf, *buf, *tmp; int ret = 0; @@ -9500,7 +9543,7 @@ ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, if (tmp) *tmp = '\0'; - ret = trace_run_command(buf, createfn); + ret = createfn(buf); if (ret) goto out; buf += size; diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h index e448d2da0b99..a6446c03cfbc 100644 --- a/kernel/trace/trace.h +++ b/kernel/trace/trace.h @@ -136,25 +136,6 @@ struct kretprobe_trace_entry_head { unsigned long ret_ip; }; -/* - * trace_flag_type is an enumeration that holds different - * states when a trace occurs. These are: - * IRQS_OFF - interrupts were disabled - * IRQS_NOSUPPORT - arch does not support irqs_disabled_flags - * NEED_RESCHED - reschedule is requested - * HARDIRQ - inside an interrupt handler - * SOFTIRQ - inside a softirq handler - */ -enum trace_flag_type { - TRACE_FLAG_IRQS_OFF = 0x01, - TRACE_FLAG_IRQS_NOSUPPORT = 0x02, - TRACE_FLAG_NEED_RESCHED = 0x04, - TRACE_FLAG_HARDIRQ = 0x08, - TRACE_FLAG_SOFTIRQ = 0x10, - TRACE_FLAG_PREEMPT_RESCHED = 0x20, - TRACE_FLAG_NMI = 0x40, -}; - #define TRACE_BUF_SIZE 1024 struct trace_array; @@ -589,8 +570,7 @@ struct ring_buffer_event * trace_buffer_lock_reserve(struct trace_buffer *buffer, int type, unsigned long len, - unsigned long flags, - int pc); + unsigned int trace_ctx); struct trace_entry *tracing_get_trace_entry(struct trace_array *tr, struct trace_array_cpu *data); @@ -601,6 +581,8 @@ struct trace_entry *trace_find_next_entry(struct trace_iterator *iter, void trace_buffer_unlock_commit_nostack(struct trace_buffer *buffer, struct ring_buffer_event *event); +const char *trace_event_format(struct trace_iterator *iter, const char *fmt); + int trace_empty(struct trace_iterator *iter); void *trace_find_next_entry_inc(struct trace_iterator *iter); @@ -615,15 +597,14 @@ unsigned long trace_total_entries(struct trace_array *tr); void trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, - unsigned long flags, int pc); + unsigned int trace_ctx); void trace_graph_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, - unsigned long flags, int pc); + unsigned int trace_ctx); void trace_latency_header(struct seq_file *m); void trace_default_header(struct seq_file *m); void print_trace_header(struct seq_file *m, struct trace_iterator *iter); -int trace_empty(struct trace_iterator *iter); void trace_graph_return(struct ftrace_graph_ret *trace); int trace_graph_entry(struct ftrace_graph_ent *trace); @@ -687,11 +668,10 @@ static inline void latency_fsnotify(struct trace_array *tr) { } #endif #ifdef CONFIG_STACKTRACE -void __trace_stack(struct trace_array *tr, unsigned long flags, int skip, - int pc); +void __trace_stack(struct trace_array *tr, unsigned int trace_ctx, int skip); #else -static inline void __trace_stack(struct trace_array *tr, unsigned long flags, - int skip, int pc) +static inline void __trace_stack(struct trace_array *tr, unsigned int trace_ctx, + int skip) { } #endif /* CONFIG_STACKTRACE */ @@ -831,10 +811,10 @@ extern void graph_trace_open(struct trace_iterator *iter); extern void graph_trace_close(struct trace_iterator *iter); extern int __trace_graph_entry(struct trace_array *tr, struct ftrace_graph_ent *trace, - unsigned long flags, int pc); + unsigned int trace_ctx); extern void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, - unsigned long flags, int pc); + unsigned int trace_ctx); #ifdef CONFIG_DYNAMIC_FTRACE extern struct ftrace_hash __rcu *ftrace_graph_hash; @@ -1194,6 +1174,7 @@ extern int trace_get_user(struct trace_parser *parser, const char __user *ubuf, C(MARKERS, "markers"), \ C(EVENT_FORK, "event-fork"), \ C(PAUSE_ON_TRACE, "pause-on-trace"), \ + C(HASH_PTR, "hash-ptr"), /* Print hashed pointer */ \ FUNCTION_FLAGS \ FGRAPH_FLAGS \ STACK_FLAGS \ @@ -1297,15 +1278,15 @@ extern int call_filter_check_discard(struct trace_event_call *call, void *rec, void trace_buffer_unlock_commit_regs(struct trace_array *tr, struct trace_buffer *buffer, struct ring_buffer_event *event, - unsigned long flags, int pc, + unsigned int trcace_ctx, struct pt_regs *regs); static inline void trace_buffer_unlock_commit(struct trace_array *tr, struct trace_buffer *buffer, struct ring_buffer_event *event, - unsigned long flags, int pc) + unsigned int trace_ctx) { - trace_buffer_unlock_commit_regs(tr, buffer, event, flags, pc, NULL); + trace_buffer_unlock_commit_regs(tr, buffer, event, trace_ctx, NULL); } DECLARE_PER_CPU(struct ring_buffer_event *, trace_buffered_event); @@ -1366,8 +1347,7 @@ __event_trigger_test_discard(struct trace_event_file *file, * @buffer: The ring buffer that the event is being written to * @event: The event meta data in the ring buffer * @entry: The event itself - * @irq_flags: The state of the interrupts at the start of the event - * @pc: The state of the preempt count at the start of the event. + * @trace_ctx: The tracing context flags. * * This is a helper function to handle triggers that require data * from the event itself. It also tests the event against filters and @@ -1377,12 +1357,12 @@ static inline void event_trigger_unlock_commit(struct trace_event_file *file, struct trace_buffer *buffer, struct ring_buffer_event *event, - void *entry, unsigned long irq_flags, int pc) + void *entry, unsigned int trace_ctx) { enum event_trigger_type tt = ETT_NONE; if (!__event_trigger_test_discard(file, buffer, event, entry, &tt)) - trace_buffer_unlock_commit(file->tr, buffer, event, irq_flags, pc); + trace_buffer_unlock_commit(file->tr, buffer, event, trace_ctx); if (tt) event_triggers_post_call(file, tt); @@ -1394,8 +1374,7 @@ event_trigger_unlock_commit(struct trace_event_file *file, * @buffer: The ring buffer that the event is being written to * @event: The event meta data in the ring buffer * @entry: The event itself - * @irq_flags: The state of the interrupts at the start of the event - * @pc: The state of the preempt count at the start of the event. + * @trace_ctx: The tracing context flags. * * This is a helper function to handle triggers that require data * from the event itself. It also tests the event against filters and @@ -1408,14 +1387,14 @@ static inline void event_trigger_unlock_commit_regs(struct trace_event_file *file, struct trace_buffer *buffer, struct ring_buffer_event *event, - void *entry, unsigned long irq_flags, int pc, + void *entry, unsigned int trace_ctx, struct pt_regs *regs) { enum event_trigger_type tt = ETT_NONE; if (!__event_trigger_test_discard(file, buffer, event, entry, &tt)) trace_buffer_unlock_commit_regs(file->tr, buffer, event, - irq_flags, pc, regs); + trace_ctx, regs); if (tt) event_triggers_post_call(file, tt); @@ -1830,10 +1809,9 @@ extern int tracing_set_cpumask(struct trace_array *tr, #define MAX_EVENT_NAME_LEN 64 -extern int trace_run_command(const char *buf, int (*createfn)(int, char**)); extern ssize_t trace_parse_run_command(struct file *file, const char __user *buffer, size_t count, loff_t *ppos, - int (*createfn)(int, char**)); + int (*createfn)(const char *)); extern unsigned int err_pos(char *cmd, const char *str); extern void tracing_log_err(struct trace_array *tr, diff --git a/kernel/trace/trace_branch.c b/kernel/trace/trace_branch.c index eff099123aa2..e47fdb4c92fb 100644 --- a/kernel/trace/trace_branch.c +++ b/kernel/trace/trace_branch.c @@ -37,7 +37,7 @@ probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) struct ring_buffer_event *event; struct trace_branch *entry; unsigned long flags; - int pc; + unsigned int trace_ctx; const char *p; if (current->trace_recursion & TRACE_BRANCH_BIT) @@ -59,10 +59,10 @@ probe_likely_condition(struct ftrace_likely_data *f, int val, int expect) if (atomic_read(&data->disabled)) goto out; - pc = preempt_count(); + trace_ctx = tracing_gen_ctx_flags(flags); buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, TRACE_BRANCH, - sizeof(*entry), flags, pc); + sizeof(*entry), trace_ctx); if (!event) goto out; diff --git a/kernel/trace/trace_dynevent.c b/kernel/trace/trace_dynevent.c index 4f967d5cd917..dc971a68dda4 100644 --- a/kernel/trace/trace_dynevent.c +++ b/kernel/trace/trace_dynevent.c @@ -31,23 +31,31 @@ int dyn_event_register(struct dyn_event_operations *ops) return 0; } -int dyn_event_release(int argc, char **argv, struct dyn_event_operations *type) +int dyn_event_release(const char *raw_command, struct dyn_event_operations *type) { struct dyn_event *pos, *n; char *system = NULL, *event, *p; - int ret = -ENOENT; + int argc, ret = -ENOENT; + char **argv; + + argv = argv_split(GFP_KERNEL, raw_command, &argc); + if (!argv) + return -ENOMEM; if (argv[0][0] == '-') { - if (argv[0][1] != ':') - return -EINVAL; + if (argv[0][1] != ':') { + ret = -EINVAL; + goto out; + } event = &argv[0][2]; } else { event = strchr(argv[0], ':'); - if (!event) - return -EINVAL; + if (!event) { + ret = -EINVAL; + goto out; + } event++; } - argc--; argv++; p = strchr(event, '/'); if (p) { @@ -63,7 +71,7 @@ int dyn_event_release(int argc, char **argv, struct dyn_event_operations *type) if (type && type != pos->ops) continue; if (!pos->ops->match(system, event, - argc, (const char **)argv, pos)) + argc - 1, (const char **)argv + 1, pos)) continue; ret = pos->ops->free(pos); @@ -71,21 +79,22 @@ int dyn_event_release(int argc, char **argv, struct dyn_event_operations *type) break; } mutex_unlock(&event_mutex); - +out: + argv_free(argv); return ret; } -static int create_dyn_event(int argc, char **argv) +static int create_dyn_event(const char *raw_command) { struct dyn_event_operations *ops; int ret = -ENODEV; - if (argv[0][0] == '-' || argv[0][0] == '!') - return dyn_event_release(argc, argv, NULL); + if (raw_command[0] == '-' || raw_command[0] == '!') + return dyn_event_release(raw_command, NULL); mutex_lock(&dyn_event_ops_mutex); list_for_each_entry(ops, &dyn_event_ops_list, list) { - ret = ops->create(argc, (const char **)argv); + ret = ops->create(raw_command); if (!ret || ret != -ECANCELED) break; } diff --git a/kernel/trace/trace_dynevent.h b/kernel/trace/trace_dynevent.h index d6f72dcb7269..7754936b57ee 100644 --- a/kernel/trace/trace_dynevent.h +++ b/kernel/trace/trace_dynevent.h @@ -39,7 +39,7 @@ struct dyn_event; */ struct dyn_event_operations { struct list_head list; - int (*create)(int argc, const char *argv[]); + int (*create)(const char *raw_command); int (*show)(struct seq_file *m, struct dyn_event *ev); bool (*is_busy)(struct dyn_event *ev); int (*free)(struct dyn_event *ev); @@ -97,7 +97,7 @@ void *dyn_event_seq_start(struct seq_file *m, loff_t *pos); void *dyn_event_seq_next(struct seq_file *m, void *v, loff_t *pos); void dyn_event_seq_stop(struct seq_file *m, void *v); int dyn_events_release_all(struct dyn_event_operations *type); -int dyn_event_release(int argc, char **argv, struct dyn_event_operations *type); +int dyn_event_release(const char *raw_command, struct dyn_event_operations *type); /* * for_each_dyn_event - iterate over the dyn_event list diff --git a/kernel/trace/trace_event_perf.c b/kernel/trace/trace_event_perf.c index a71181655958..288ad2c274fb 100644 --- a/kernel/trace/trace_event_perf.c +++ b/kernel/trace/trace_event_perf.c @@ -421,11 +421,8 @@ NOKPROBE_SYMBOL(perf_trace_buf_alloc); void perf_trace_buf_update(void *record, u16 type) { struct trace_entry *entry = record; - int pc = preempt_count(); - unsigned long flags; - local_save_flags(flags); - tracing_generic_entry_update(entry, type, flags, pc); + tracing_generic_entry_update(entry, type, tracing_gen_ctx()); } NOKPROBE_SYMBOL(perf_trace_buf_update); diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c index d387b774ceeb..a3563afd412d 100644 --- a/kernel/trace/trace_events.c +++ b/kernel/trace/trace_events.c @@ -258,22 +258,19 @@ void *trace_event_buffer_reserve(struct trace_event_buffer *fbuffer, trace_event_ignore_this_pid(trace_file)) return NULL; - local_save_flags(fbuffer->flags); - fbuffer->pc = preempt_count(); /* * If CONFIG_PREEMPTION is enabled, then the tracepoint itself disables * preemption (adding one to the preempt_count). Since we are * interested in the preempt_count at the time the tracepoint was * hit, we need to subtract one to offset the increment. */ - if (IS_ENABLED(CONFIG_PREEMPTION)) - fbuffer->pc--; + fbuffer->trace_ctx = tracing_gen_ctx_dec(); fbuffer->trace_file = trace_file; fbuffer->event = trace_event_buffer_lock_reserve(&fbuffer->buffer, trace_file, event_call->event.type, len, - fbuffer->flags, fbuffer->pc); + fbuffer->trace_ctx); if (!fbuffer->event) return NULL; @@ -2101,16 +2098,20 @@ event_subsystem_dir(struct trace_array *tr, const char *name, dir->subsystem = system; file->system = dir; - entry = tracefs_create_file("filter", 0644, dir->entry, dir, - &ftrace_subsystem_filter_fops); - if (!entry) { - kfree(system->filter); - system->filter = NULL; - pr_warn("Could not create tracefs '%s/filter' entry\n", name); - } + /* the ftrace system is special, do not create enable or filter files */ + if (strcmp(name, "ftrace") != 0) { - trace_create_file("enable", 0644, dir->entry, dir, - &ftrace_system_enable_fops); + entry = tracefs_create_file("filter", 0644, dir->entry, dir, + &ftrace_subsystem_filter_fops); + if (!entry) { + kfree(system->filter); + system->filter = NULL; + pr_warn("Could not create tracefs '%s/filter' entry\n", name); + } + + trace_create_file("enable", 0644, dir->entry, dir, + &ftrace_system_enable_fops); + } list_add(&dir->list, &tr->systems); @@ -3679,12 +3680,11 @@ function_test_events_call(unsigned long ip, unsigned long parent_ip, struct trace_buffer *buffer; struct ring_buffer_event *event; struct ftrace_entry *entry; - unsigned long flags; + unsigned int trace_ctx; long disabled; int cpu; - int pc; - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); preempt_disable_notrace(); cpu = raw_smp_processor_id(); disabled = atomic_inc_return(&per_cpu(ftrace_test_event_disable, cpu)); @@ -3692,11 +3692,9 @@ function_test_events_call(unsigned long ip, unsigned long parent_ip, if (disabled != 1) goto out; - local_save_flags(flags); - event = trace_event_buffer_lock_reserve(&buffer, &event_trace_file, TRACE_FN, sizeof(*entry), - flags, pc); + trace_ctx); if (!event) goto out; entry = ring_buffer_event_data(event); @@ -3704,7 +3702,7 @@ function_test_events_call(unsigned long ip, unsigned long parent_ip, entry->parent_ip = parent_ip; event_trigger_unlock_commit(&event_trace_file, buffer, event, - entry, flags, pc); + entry, trace_ctx); out: atomic_dec(&per_cpu(ftrace_test_event_disable, cpu)); preempt_enable_notrace(); diff --git a/kernel/trace/trace_events_inject.c b/kernel/trace/trace_events_inject.c index 22bcf7c51d1e..c188045c5f97 100644 --- a/kernel/trace/trace_events_inject.c +++ b/kernel/trace/trace_events_inject.c @@ -192,7 +192,6 @@ static void *trace_alloc_entry(struct trace_event_call *call, int *size) static int parse_entry(char *str, struct trace_event_call *call, void **pentry) { struct ftrace_event_field *field; - unsigned long irq_flags; void *entry = NULL; int entry_size; u64 val = 0; @@ -203,9 +202,8 @@ static int parse_entry(char *str, struct trace_event_call *call, void **pentry) if (!entry) return -ENOMEM; - local_save_flags(irq_flags); - tracing_generic_entry_update(entry, call->event.type, irq_flags, - preempt_count()); + tracing_generic_entry_update(entry, call->event.type, + tracing_gen_ctx()); while ((len = parse_field(str, call, &field, &val)) > 0) { if (is_function_field(field)) diff --git a/kernel/trace/trace_events_synth.c b/kernel/trace/trace_events_synth.c index 5a8bc0b421f1..8d71e6c83f10 100644 --- a/kernel/trace/trace_events_synth.c +++ b/kernel/trace/trace_events_synth.c @@ -23,13 +23,14 @@ #undef ERRORS #define ERRORS \ C(BAD_NAME, "Illegal name"), \ - C(CMD_INCOMPLETE, "Incomplete command"), \ + C(INVALID_CMD, "Command must be of the form: <name> field[;field] ..."),\ + C(INVALID_DYN_CMD, "Command must be of the form: s or -:[synthetic/]<name> field[;field] ..."),\ C(EVENT_EXISTS, "Event already exists"), \ C(TOO_MANY_FIELDS, "Too many fields"), \ C(INCOMPLETE_TYPE, "Incomplete type"), \ C(INVALID_TYPE, "Invalid type"), \ - C(INVALID_FIELD, "Invalid field"), \ - C(CMD_TOO_LONG, "Command too long"), + C(INVALID_FIELD, "Invalid field"), \ + C(INVALID_ARRAY_SPEC, "Invalid array specification"), #undef C #define C(a, b) SYNTH_ERR_##a @@ -48,7 +49,7 @@ static int errpos(const char *str) return err_pos(last_cmd, str); } -static void last_cmd_set(char *str) +static void last_cmd_set(const char *str) { if (!str) return; @@ -62,7 +63,7 @@ static void synth_err(u8 err_type, u8 err_pos) err_type, err_pos); } -static int create_synth_event(int argc, const char **argv); +static int create_synth_event(const char *raw_command); static int synth_event_show(struct seq_file *m, struct dyn_event *ev); static int synth_event_release(struct dyn_event *ev); static bool synth_event_is_busy(struct dyn_event *ev); @@ -579,18 +580,32 @@ static void free_synth_field(struct synth_field *field) kfree(field); } -static struct synth_field *parse_synth_field(int argc, const char **argv, - int *consumed) +static int check_field_version(const char *prefix, const char *field_type, + const char *field_name) +{ + /* + * For backward compatibility, the old synthetic event command + * format did not require semicolons, and in order to not + * break user space, that old format must still work. If a new + * feature is added, then the format that uses the new feature + * will be required to have semicolons, as nothing that uses + * the old format would be using the new, yet to be created, + * feature. When a new feature is added, this will detect it, + * and return a number greater than 1, and require the format + * to use semicolons. + */ + return 1; +} + +static struct synth_field *parse_synth_field(int argc, char **argv, + int *consumed, int *field_version) { - struct synth_field *field; const char *prefix = NULL, *field_type = argv[0], *field_name, *array; + struct synth_field *field; int len, ret = -ENOMEM; struct seq_buf s; ssize_t size; - if (field_type[0] == ';') - field_type++; - if (!strcmp(field_type, "unsigned")) { if (argc < 3) { synth_err(SYNTH_ERR_INCOMPLETE_TYPE, errpos(field_type)); @@ -599,12 +614,19 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, prefix = "unsigned "; field_type = argv[1]; field_name = argv[2]; - *consumed = 3; + *consumed += 3; } else { field_name = argv[1]; - *consumed = 2; + *consumed += 2; } + if (!field_name) { + synth_err(SYNTH_ERR_INVALID_FIELD, errpos(field_type)); + return ERR_PTR(-EINVAL); + } + + *field_version = check_field_version(prefix, field_type, field_name); + field = kzalloc(sizeof(*field), GFP_KERNEL); if (!field) return ERR_PTR(-ENOMEM); @@ -613,8 +635,6 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, array = strchr(field_name, '['); if (array) len -= strlen(array); - else if (field_name[len - 1] == ';') - len--; field->name = kmemdup_nul(field_name, len, GFP_KERNEL); if (!field->name) @@ -626,8 +646,6 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, goto free; } - if (field_type[0] == ';') - field_type++; len = strlen(field_type) + 1; if (array) @@ -644,11 +662,8 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, if (prefix) seq_buf_puts(&s, prefix); seq_buf_puts(&s, field_type); - if (array) { + if (array) seq_buf_puts(&s, array); - if (s.buffer[s.len - 1] == ';') - s.len--; - } if (WARN_ON_ONCE(!seq_buf_buffer_left(&s))) goto free; @@ -656,7 +671,10 @@ static struct synth_field *parse_synth_field(int argc, const char **argv, size = synth_field_size(field->type); if (size < 0) { - synth_err(SYNTH_ERR_INVALID_TYPE, errpos(field_type)); + if (array) + synth_err(SYNTH_ERR_INVALID_ARRAY_SPEC, errpos(field_name)); + else + synth_err(SYNTH_ERR_INVALID_TYPE, errpos(field_type)); ret = -EINVAL; goto free; } else if (size == 0) { @@ -1160,46 +1178,13 @@ int synth_event_gen_cmd_array_start(struct dynevent_cmd *cmd, const char *name, } EXPORT_SYMBOL_GPL(synth_event_gen_cmd_array_start); -static int save_cmdstr(int argc, const char *name, const char **argv) -{ - struct seq_buf s; - char *buf; - int i; - - buf = kzalloc(MAX_DYNEVENT_CMD_LEN, GFP_KERNEL); - if (!buf) - return -ENOMEM; - - seq_buf_init(&s, buf, MAX_DYNEVENT_CMD_LEN); - - seq_buf_puts(&s, name); - - for (i = 0; i < argc; i++) { - seq_buf_putc(&s, ' '); - seq_buf_puts(&s, argv[i]); - } - - if (!seq_buf_buffer_left(&s)) { - synth_err(SYNTH_ERR_CMD_TOO_LONG, 0); - kfree(buf); - return -EINVAL; - } - buf[s.len] = 0; - last_cmd_set(buf); - - kfree(buf); - return 0; -} - -static int __create_synth_event(int argc, const char *name, const char **argv) +static int __create_synth_event(const char *name, const char *raw_fields) { + char **argv, *field_str, *tmp_fields, *saved_fields = NULL; struct synth_field *field, *fields[SYNTH_FIELDS_MAX]; + int consumed, cmd_version = 1, n_fields_this_loop; + int i, argc, n_fields = 0, ret = 0; struct synth_event *event = NULL; - int i, consumed = 0, n_fields = 0, ret = 0; - - ret = save_cmdstr(argc, name, argv); - if (ret) - return ret; /* * Argument syntax: @@ -1208,46 +1193,101 @@ static int __create_synth_event(int argc, const char *name, const char **argv) * where 'field' = type field_name */ - if (name[0] == '\0' || argc < 1) { - synth_err(SYNTH_ERR_CMD_INCOMPLETE, 0); + if (name[0] == '\0') { + synth_err(SYNTH_ERR_INVALID_CMD, 0); return -EINVAL; } - mutex_lock(&event_mutex); - if (!is_good_name(name)) { synth_err(SYNTH_ERR_BAD_NAME, errpos(name)); - ret = -EINVAL; - goto out; + return -EINVAL; } + mutex_lock(&event_mutex); + event = find_synth_event(name); if (event) { synth_err(SYNTH_ERR_EVENT_EXISTS, errpos(name)); ret = -EEXIST; - goto out; + goto err; } - for (i = 0; i < argc - 1; i++) { - if (strcmp(argv[i], ";") == 0) - continue; - if (n_fields == SYNTH_FIELDS_MAX) { - synth_err(SYNTH_ERR_TOO_MANY_FIELDS, 0); - ret = -EINVAL; + tmp_fields = saved_fields = kstrdup(raw_fields, GFP_KERNEL); + if (!tmp_fields) { + ret = -ENOMEM; + goto err; + } + + while ((field_str = strsep(&tmp_fields, ";")) != NULL) { + argv = argv_split(GFP_KERNEL, field_str, &argc); + if (!argv) { + ret = -ENOMEM; goto err; } - field = parse_synth_field(argc - i, &argv[i], &consumed); - if (IS_ERR(field)) { - ret = PTR_ERR(field); + if (!argc) { + argv_free(argv); + continue; + } + + n_fields_this_loop = 0; + consumed = 0; + while (argc > consumed) { + int field_version; + + field = parse_synth_field(argc - consumed, + argv + consumed, &consumed, + &field_version); + if (IS_ERR(field)) { + argv_free(argv); + ret = PTR_ERR(field); + goto err; + } + + /* + * Track the highest version of any field we + * found in the command. + */ + if (field_version > cmd_version) + cmd_version = field_version; + + /* + * Now sort out what is and isn't valid for + * each supported version. + * + * If we see more than 1 field per loop, it + * means we have multiple fields between + * semicolons, and that's something we no + * longer support in a version 2 or greater + * command. + */ + if (cmd_version > 1 && n_fields_this_loop >= 1) { + synth_err(SYNTH_ERR_INVALID_CMD, errpos(field_str)); + ret = -EINVAL; + goto err; + } + + fields[n_fields++] = field; + if (n_fields == SYNTH_FIELDS_MAX) { + synth_err(SYNTH_ERR_TOO_MANY_FIELDS, 0); + ret = -EINVAL; + goto err; + } + + n_fields_this_loop++; + } + + if (consumed < argc) { + synth_err(SYNTH_ERR_INVALID_CMD, 0); + ret = -EINVAL; goto err; } - fields[n_fields++] = field; - i += consumed - 1; + + argv_free(argv); } - if (i < argc && strcmp(argv[i], ";") != 0) { - synth_err(SYNTH_ERR_INVALID_FIELD, errpos(argv[i])); + if (n_fields == 0) { + synth_err(SYNTH_ERR_INVALID_CMD, 0); ret = -EINVAL; goto err; } @@ -1266,6 +1306,8 @@ static int __create_synth_event(int argc, const char *name, const char **argv) out: mutex_unlock(&event_mutex); + kfree(saved_fields); + return ret; err: for (i = 0; i < n_fields; i++) @@ -1383,19 +1425,79 @@ int synth_event_delete(const char *event_name) } EXPORT_SYMBOL_GPL(synth_event_delete); -static int create_or_delete_synth_event(int argc, char **argv) +static int check_command(const char *raw_command) { - const char *name = argv[0]; - int ret; + char **argv = NULL, *cmd, *saved_cmd, *name_and_field; + int argc, ret = 0; + + cmd = saved_cmd = kstrdup(raw_command, GFP_KERNEL); + if (!cmd) + return -ENOMEM; + + name_and_field = strsep(&cmd, ";"); + if (!name_and_field) { + ret = -EINVAL; + goto free; + } + + if (name_and_field[0] == '!') + goto free; + + argv = argv_split(GFP_KERNEL, name_and_field, &argc); + if (!argv) { + ret = -ENOMEM; + goto free; + } + argv_free(argv); + + if (argc < 3) + ret = -EINVAL; +free: + kfree(saved_cmd); + + return ret; +} + +static int create_or_delete_synth_event(const char *raw_command) +{ + char *name = NULL, *fields, *p; + int ret = 0; + + raw_command = skip_spaces(raw_command); + if (raw_command[0] == '\0') + return ret; + + last_cmd_set(raw_command); + + ret = check_command(raw_command); + if (ret) { + synth_err(SYNTH_ERR_INVALID_CMD, 0); + return ret; + } + + p = strpbrk(raw_command, " \t"); + if (!p && raw_command[0] != '!') { + synth_err(SYNTH_ERR_INVALID_CMD, 0); + ret = -EINVAL; + goto free; + } + + name = kmemdup_nul(raw_command, p ? p - raw_command : strlen(raw_command), GFP_KERNEL); + if (!name) + return -ENOMEM; - /* trace_run_command() ensures argc != 0 */ if (name[0] == '!') { ret = synth_event_delete(name + 1); - return ret; + goto free; } - ret = __create_synth_event(argc - 1, name, (const char **)argv + 1); - return ret == -ECANCELED ? -EINVAL : ret; + fields = skip_spaces(p); + + ret = __create_synth_event(name, fields); +free: + kfree(name); + + return ret; } static int synth_event_run_command(struct dynevent_cmd *cmd) @@ -1403,7 +1505,7 @@ static int synth_event_run_command(struct dynevent_cmd *cmd) struct synth_event *se; int ret; - ret = trace_run_command(cmd->seq.buffer, create_or_delete_synth_event); + ret = create_or_delete_synth_event(cmd->seq.buffer); if (ret) return ret; @@ -1939,10 +2041,27 @@ int synth_event_trace_end(struct synth_event_trace_state *trace_state) } EXPORT_SYMBOL_GPL(synth_event_trace_end); -static int create_synth_event(int argc, const char **argv) +static int create_synth_event(const char *raw_command) { - const char *name = argv[0]; - int len; + char *fields, *p; + const char *name; + int len, ret = 0; + + raw_command = skip_spaces(raw_command); + if (raw_command[0] == '\0') + return ret; + + last_cmd_set(raw_command); + + p = strpbrk(raw_command, " \t"); + if (!p) { + synth_err(SYNTH_ERR_INVALID_CMD, 0); + return -EINVAL; + } + + fields = skip_spaces(p); + + name = raw_command; if (name[0] != 's' || name[1] != ':') return -ECANCELED; @@ -1951,11 +2070,30 @@ static int create_synth_event(int argc, const char **argv) /* This interface accepts group name prefix */ if (strchr(name, '/')) { len = str_has_prefix(name, SYNTH_SYSTEM "/"); - if (len == 0) + if (len == 0) { + synth_err(SYNTH_ERR_INVALID_DYN_CMD, 0); return -EINVAL; + } name += len; } - return __create_synth_event(argc - 1, name, argv + 1); + + len = name - raw_command; + + ret = check_command(raw_command + len); + if (ret) { + synth_err(SYNTH_ERR_INVALID_CMD, 0); + return ret; + } + + name = kmemdup_nul(raw_command + len, p - raw_command - len, GFP_KERNEL); + if (!name) + return -ENOMEM; + + ret = __create_synth_event(name, fields); + + kfree(name); + + return ret; } static int synth_event_release(struct dyn_event *ev) diff --git a/kernel/trace/trace_functions.c b/kernel/trace/trace_functions.c index c5095dd28e20..f93723ca66bc 100644 --- a/kernel/trace/trace_functions.c +++ b/kernel/trace/trace_functions.c @@ -106,8 +106,7 @@ static int function_trace_init(struct trace_array *tr) ftrace_init_array_ops(tr, func); - tr->array_buffer.cpu = get_cpu(); - put_cpu(); + tr->array_buffer.cpu = raw_smp_processor_id(); tracing_start_cmdline_record(); tracing_start_function_trace(tr); @@ -132,10 +131,9 @@ function_trace_call(unsigned long ip, unsigned long parent_ip, { struct trace_array *tr = op->private; struct trace_array_cpu *data; - unsigned long flags; + unsigned int trace_ctx; int bit; int cpu; - int pc; if (unlikely(!tr->function_enabled)) return; @@ -144,15 +142,14 @@ function_trace_call(unsigned long ip, unsigned long parent_ip, if (bit < 0) return; - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); preempt_disable_notrace(); cpu = smp_processor_id(); data = per_cpu_ptr(tr->array_buffer.data, cpu); - if (!atomic_read(&data->disabled)) { - local_save_flags(flags); - trace_function(tr, ip, parent_ip, flags, pc); - } + if (!atomic_read(&data->disabled)) + trace_function(tr, ip, parent_ip, trace_ctx); + ftrace_test_recursion_unlock(bit); preempt_enable_notrace(); } @@ -184,7 +181,7 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, unsigned long flags; long disabled; int cpu; - int pc; + unsigned int trace_ctx; if (unlikely(!tr->function_enabled)) return; @@ -199,9 +196,9 @@ function_stack_trace_call(unsigned long ip, unsigned long parent_ip, disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { - pc = preempt_count(); - trace_function(tr, ip, parent_ip, flags, pc); - __trace_stack(tr, flags, STACK_SKIP, pc); + trace_ctx = tracing_gen_ctx_flags(flags); + trace_function(tr, ip, parent_ip, trace_ctx); + __trace_stack(tr, trace_ctx, STACK_SKIP); } atomic_dec(&data->disabled); @@ -404,13 +401,11 @@ ftrace_traceoff(unsigned long ip, unsigned long parent_ip, static __always_inline void trace_stack(struct trace_array *tr) { - unsigned long flags; - int pc; + unsigned int trace_ctx; - local_save_flags(flags); - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); - __trace_stack(tr, flags, FTRACE_STACK_SKIP, pc); + __trace_stack(tr, trace_ctx, FTRACE_STACK_SKIP); } static void diff --git a/kernel/trace/trace_functions_graph.c b/kernel/trace/trace_functions_graph.c index d874dec87131..0aa6e6faa943 100644 --- a/kernel/trace/trace_functions_graph.c +++ b/kernel/trace/trace_functions_graph.c @@ -96,8 +96,7 @@ print_graph_duration(struct trace_array *tr, unsigned long long duration, int __trace_graph_entry(struct trace_array *tr, struct ftrace_graph_ent *trace, - unsigned long flags, - int pc) + unsigned int trace_ctx) { struct trace_event_call *call = &event_funcgraph_entry; struct ring_buffer_event *event; @@ -105,7 +104,7 @@ int __trace_graph_entry(struct trace_array *tr, struct ftrace_graph_ent_entry *entry; event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_ENT, - sizeof(*entry), flags, pc); + sizeof(*entry), trace_ctx); if (!event) return 0; entry = ring_buffer_event_data(event); @@ -129,10 +128,10 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) struct trace_array *tr = graph_array; struct trace_array_cpu *data; unsigned long flags; + unsigned int trace_ctx; long disabled; int ret; int cpu; - int pc; if (trace_recursion_test(TRACE_GRAPH_NOTRACE_BIT)) return 0; @@ -174,8 +173,8 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { - pc = preempt_count(); - ret = __trace_graph_entry(tr, trace, flags, pc); + trace_ctx = tracing_gen_ctx_flags(flags); + ret = __trace_graph_entry(tr, trace, trace_ctx); } else { ret = 0; } @@ -188,7 +187,7 @@ int trace_graph_entry(struct ftrace_graph_ent *trace) static void __trace_graph_function(struct trace_array *tr, - unsigned long ip, unsigned long flags, int pc) + unsigned long ip, unsigned int trace_ctx) { u64 time = trace_clock_local(); struct ftrace_graph_ent ent = { @@ -202,22 +201,21 @@ __trace_graph_function(struct trace_array *tr, .rettime = time, }; - __trace_graph_entry(tr, &ent, flags, pc); - __trace_graph_return(tr, &ret, flags, pc); + __trace_graph_entry(tr, &ent, trace_ctx); + __trace_graph_return(tr, &ret, trace_ctx); } void trace_graph_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, - unsigned long flags, int pc) + unsigned int trace_ctx) { - __trace_graph_function(tr, ip, flags, pc); + __trace_graph_function(tr, ip, trace_ctx); } void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret *trace, - unsigned long flags, - int pc) + unsigned int trace_ctx) { struct trace_event_call *call = &event_funcgraph_exit; struct ring_buffer_event *event; @@ -225,7 +223,7 @@ void __trace_graph_return(struct trace_array *tr, struct ftrace_graph_ret_entry *entry; event = trace_buffer_lock_reserve(buffer, TRACE_GRAPH_RET, - sizeof(*entry), flags, pc); + sizeof(*entry), trace_ctx); if (!event) return; entry = ring_buffer_event_data(event); @@ -239,9 +237,9 @@ void trace_graph_return(struct ftrace_graph_ret *trace) struct trace_array *tr = graph_array; struct trace_array_cpu *data; unsigned long flags; + unsigned int trace_ctx; long disabled; int cpu; - int pc; ftrace_graph_addr_finish(trace); @@ -255,8 +253,8 @@ void trace_graph_return(struct ftrace_graph_ret *trace) data = per_cpu_ptr(tr->array_buffer.data, cpu); disabled = atomic_inc_return(&data->disabled); if (likely(disabled == 1)) { - pc = preempt_count(); - __trace_graph_return(tr, trace, flags, pc); + trace_ctx = tracing_gen_ctx_flags(flags); + __trace_graph_return(tr, trace, trace_ctx); } atomic_dec(&data->disabled); local_irq_restore(flags); diff --git a/kernel/trace/trace_hwlat.c b/kernel/trace/trace_hwlat.c index c0df9b97f147..34dc1a712dcb 100644 --- a/kernel/trace/trace_hwlat.c +++ b/kernel/trace/trace_hwlat.c @@ -108,14 +108,9 @@ static void trace_hwlat_sample(struct hwlat_sample *sample) struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct hwlat_entry *entry; - unsigned long flags; - int pc; - - pc = preempt_count(); - local_save_flags(flags); event = trace_buffer_lock_reserve(buffer, TRACE_HWLAT, sizeof(*entry), - flags, pc); + tracing_gen_ctx()); if (!event) return; entry = ring_buffer_event_data(event); diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c index 6756379b661f..590b3d51afae 100644 --- a/kernel/trace/trace_irqsoff.c +++ b/kernel/trace/trace_irqsoff.c @@ -143,11 +143,14 @@ irqsoff_tracer_call(unsigned long ip, unsigned long parent_ip, struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; + unsigned int trace_ctx; if (!func_prolog_dec(tr, &data, &flags)) return; - trace_function(tr, ip, parent_ip, flags, preempt_count()); + trace_ctx = tracing_gen_ctx_flags(flags); + + trace_function(tr, ip, parent_ip, trace_ctx); atomic_dec(&data->disabled); } @@ -177,8 +180,8 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; + unsigned int trace_ctx; int ret; - int pc; if (ftrace_graph_ignore_func(trace)) return 0; @@ -195,8 +198,8 @@ static int irqsoff_graph_entry(struct ftrace_graph_ent *trace) if (!func_prolog_dec(tr, &data, &flags)) return 0; - pc = preempt_count(); - ret = __trace_graph_entry(tr, trace, flags, pc); + trace_ctx = tracing_gen_ctx_flags(flags); + ret = __trace_graph_entry(tr, trace, trace_ctx); atomic_dec(&data->disabled); return ret; @@ -207,15 +210,15 @@ static void irqsoff_graph_return(struct ftrace_graph_ret *trace) struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; unsigned long flags; - int pc; + unsigned int trace_ctx; ftrace_graph_addr_finish(trace); if (!func_prolog_dec(tr, &data, &flags)) return; - pc = preempt_count(); - __trace_graph_return(tr, trace, flags, pc); + trace_ctx = tracing_gen_ctx_flags(flags); + __trace_graph_return(tr, trace, trace_ctx); atomic_dec(&data->disabled); } @@ -267,12 +270,12 @@ static void irqsoff_print_header(struct seq_file *s) static void __trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, - unsigned long flags, int pc) + unsigned int trace_ctx) { if (is_graph(tr)) - trace_graph_function(tr, ip, parent_ip, flags, pc); + trace_graph_function(tr, ip, parent_ip, trace_ctx); else - trace_function(tr, ip, parent_ip, flags, pc); + trace_function(tr, ip, parent_ip, trace_ctx); } #else @@ -322,15 +325,13 @@ check_critical_timing(struct trace_array *tr, { u64 T0, T1, delta; unsigned long flags; - int pc; + unsigned int trace_ctx; T0 = data->preempt_timestamp; T1 = ftrace_now(cpu); delta = T1-T0; - local_save_flags(flags); - - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); if (!report_latency(tr, delta)) goto out; @@ -341,9 +342,9 @@ check_critical_timing(struct trace_array *tr, if (!report_latency(tr, delta)) goto out_unlock; - __trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc); + __trace_function(tr, CALLER_ADDR0, parent_ip, trace_ctx); /* Skip 5 functions to get to the irq/preempt enable function */ - __trace_stack(tr, flags, 5, pc); + __trace_stack(tr, trace_ctx, 5); if (data->critical_sequence != max_sequence) goto out_unlock; @@ -363,16 +364,15 @@ out_unlock: out: data->critical_sequence = max_sequence; data->preempt_timestamp = ftrace_now(cpu); - __trace_function(tr, CALLER_ADDR0, parent_ip, flags, pc); + __trace_function(tr, CALLER_ADDR0, parent_ip, trace_ctx); } static nokprobe_inline void -start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) +start_critical_timing(unsigned long ip, unsigned long parent_ip) { int cpu; struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; - unsigned long flags; if (!tracer_enabled || !tracing_is_enabled()) return; @@ -393,9 +393,7 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) data->preempt_timestamp = ftrace_now(cpu); data->critical_start = parent_ip ? : ip; - local_save_flags(flags); - - __trace_function(tr, ip, parent_ip, flags, pc); + __trace_function(tr, ip, parent_ip, tracing_gen_ctx()); per_cpu(tracing_cpu, cpu) = 1; @@ -403,12 +401,12 @@ start_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) } static nokprobe_inline void -stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) +stop_critical_timing(unsigned long ip, unsigned long parent_ip) { int cpu; struct trace_array *tr = irqsoff_trace; struct trace_array_cpu *data; - unsigned long flags; + unsigned int trace_ctx; cpu = raw_smp_processor_id(); /* Always clear the tracing cpu on stopping the trace */ @@ -428,8 +426,8 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) atomic_inc(&data->disabled); - local_save_flags(flags); - __trace_function(tr, ip, parent_ip, flags, pc); + trace_ctx = tracing_gen_ctx(); + __trace_function(tr, ip, parent_ip, trace_ctx); check_critical_timing(tr, data, parent_ip ? : ip, cpu); data->critical_start = 0; atomic_dec(&data->disabled); @@ -438,20 +436,16 @@ stop_critical_timing(unsigned long ip, unsigned long parent_ip, int pc) /* start and stop critical timings used to for stoppage (in idle) */ void start_critical_timings(void) { - int pc = preempt_count(); - - if (preempt_trace(pc) || irq_trace()) - start_critical_timing(CALLER_ADDR0, CALLER_ADDR1, pc); + if (preempt_trace(preempt_count()) || irq_trace()) + start_critical_timing(CALLER_ADDR0, CALLER_ADDR1); } EXPORT_SYMBOL_GPL(start_critical_timings); NOKPROBE_SYMBOL(start_critical_timings); void stop_critical_timings(void) { - int pc = preempt_count(); - - if (preempt_trace(pc) || irq_trace()) - stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1, pc); + if (preempt_trace(preempt_count()) || irq_trace()) + stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1); } EXPORT_SYMBOL_GPL(stop_critical_timings); NOKPROBE_SYMBOL(stop_critical_timings); @@ -613,19 +607,15 @@ static void irqsoff_tracer_stop(struct trace_array *tr) */ void tracer_hardirqs_on(unsigned long a0, unsigned long a1) { - unsigned int pc = preempt_count(); - - if (!preempt_trace(pc) && irq_trace()) - stop_critical_timing(a0, a1, pc); + if (!preempt_trace(preempt_count()) && irq_trace()) + stop_critical_timing(a0, a1); } NOKPROBE_SYMBOL(tracer_hardirqs_on); void tracer_hardirqs_off(unsigned long a0, unsigned long a1) { - unsigned int pc = preempt_count(); - - if (!preempt_trace(pc) && irq_trace()) - start_critical_timing(a0, a1, pc); + if (!preempt_trace(preempt_count()) && irq_trace()) + start_critical_timing(a0, a1); } NOKPROBE_SYMBOL(tracer_hardirqs_off); @@ -665,18 +655,14 @@ static struct tracer irqsoff_tracer __read_mostly = #ifdef CONFIG_PREEMPT_TRACER void tracer_preempt_on(unsigned long a0, unsigned long a1) { - int pc = preempt_count(); - - if (preempt_trace(pc) && !irq_trace()) - stop_critical_timing(a0, a1, pc); + if (preempt_trace(preempt_count()) && !irq_trace()) + stop_critical_timing(a0, a1); } void tracer_preempt_off(unsigned long a0, unsigned long a1) { - int pc = preempt_count(); - - if (preempt_trace(pc) && !irq_trace()) - start_critical_timing(a0, a1, pc); + if (preempt_trace(preempt_count()) && !irq_trace()) + start_critical_timing(a0, a1); } static int preemptoff_tracer_init(struct trace_array *tr) diff --git a/kernel/trace/trace_kprobe.c b/kernel/trace/trace_kprobe.c index 56c7fbff7bd7..6fe770d86dc3 100644 --- a/kernel/trace/trace_kprobe.c +++ b/kernel/trace/trace_kprobe.c @@ -35,7 +35,7 @@ static int __init set_kprobe_boot_events(char *str) } __setup("kprobe_event=", set_kprobe_boot_events); -static int trace_kprobe_create(int argc, const char **argv); +static int trace_kprobe_create(const char *raw_command); static int trace_kprobe_show(struct seq_file *m, struct dyn_event *ev); static int trace_kprobe_release(struct dyn_event *ev); static bool trace_kprobe_is_busy(struct dyn_event *ev); @@ -124,9 +124,9 @@ static nokprobe_inline bool trace_kprobe_module_exist(struct trace_kprobe *tk) if (!p) return true; *p = '\0'; - mutex_lock(&module_mutex); + rcu_read_lock_sched(); ret = !!find_module(tk->symbol); - mutex_unlock(&module_mutex); + rcu_read_unlock_sched(); *p = ':'; return ret; @@ -711,7 +711,7 @@ static inline void sanitize_event_name(char *name) *name = '_'; } -static int trace_kprobe_create(int argc, const char *argv[]) +static int __trace_kprobe_create(int argc, const char *argv[]) { /* * Argument syntax: @@ -910,20 +910,25 @@ error: goto out; } -static int create_or_delete_trace_kprobe(int argc, char **argv) +static int trace_kprobe_create(const char *raw_command) +{ + return trace_probe_create(raw_command, __trace_kprobe_create); +} + +static int create_or_delete_trace_kprobe(const char *raw_command) { int ret; - if (argv[0][0] == '-') - return dyn_event_release(argc, argv, &trace_kprobe_ops); + if (raw_command[0] == '-') + return dyn_event_release(raw_command, &trace_kprobe_ops); - ret = trace_kprobe_create(argc, (const char **)argv); + ret = trace_kprobe_create(raw_command); return ret == -ECANCELED ? -EINVAL : ret; } static int trace_kprobe_run_command(struct dynevent_cmd *cmd) { - return trace_run_command(cmd->seq.buffer, create_or_delete_trace_kprobe); + return create_or_delete_trace_kprobe(cmd->seq.buffer); } /** @@ -1084,7 +1089,7 @@ int kprobe_event_delete(const char *name) snprintf(buf, MAX_EVENT_NAME_LEN, "-:%s", name); - return trace_run_command(buf, create_or_delete_trace_kprobe); + return create_or_delete_trace_kprobe(buf); } EXPORT_SYMBOL_GPL(kprobe_event_delete); @@ -1386,8 +1391,7 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, if (trace_trigger_soft_disabled(trace_file)) return; - local_save_flags(fbuffer.flags); - fbuffer.pc = preempt_count(); + fbuffer.trace_ctx = tracing_gen_ctx(); fbuffer.trace_file = trace_file; dsize = __get_data_size(&tk->tp, regs); @@ -1396,7 +1400,7 @@ __kprobe_trace_func(struct trace_kprobe *tk, struct pt_regs *regs, trace_event_buffer_lock_reserve(&fbuffer.buffer, trace_file, call->event.type, sizeof(*entry) + tk->tp.size + dsize, - fbuffer.flags, fbuffer.pc); + fbuffer.trace_ctx); if (!fbuffer.event) return; @@ -1434,8 +1438,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, if (trace_trigger_soft_disabled(trace_file)) return; - local_save_flags(fbuffer.flags); - fbuffer.pc = preempt_count(); + fbuffer.trace_ctx = tracing_gen_ctx(); fbuffer.trace_file = trace_file; dsize = __get_data_size(&tk->tp, regs); @@ -1443,7 +1446,7 @@ __kretprobe_trace_func(struct trace_kprobe *tk, struct kretprobe_instance *ri, trace_event_buffer_lock_reserve(&fbuffer.buffer, trace_file, call->event.type, sizeof(*entry) + tk->tp.size + dsize, - fbuffer.flags, fbuffer.pc); + fbuffer.trace_ctx); if (!fbuffer.event) return; @@ -1888,7 +1891,7 @@ static __init void setup_boot_kprobe_events(void) if (p) *p++ = '\0'; - ret = trace_run_command(cmd, create_or_delete_trace_kprobe); + ret = create_or_delete_trace_kprobe(cmd); if (ret) pr_warn("Failed to add event(%d): %s\n", ret, cmd); @@ -1982,8 +1985,7 @@ static __init int kprobe_trace_self_tests_init(void) pr_info("Testing kprobe tracing: "); - ret = trace_run_command("p:testprobe kprobe_trace_selftest_target $stack $stack0 +0($stack)", - create_or_delete_trace_kprobe); + ret = create_or_delete_trace_kprobe("p:testprobe kprobe_trace_selftest_target $stack $stack0 +0($stack)"); if (WARN_ON_ONCE(ret)) { pr_warn("error on probing function entry.\n"); warn++; @@ -2004,8 +2006,7 @@ static __init int kprobe_trace_self_tests_init(void) } } - ret = trace_run_command("r:testprobe2 kprobe_trace_selftest_target $retval", - create_or_delete_trace_kprobe); + ret = create_or_delete_trace_kprobe("r:testprobe2 kprobe_trace_selftest_target $retval"); if (WARN_ON_ONCE(ret)) { pr_warn("error on probing function return.\n"); warn++; @@ -2078,13 +2079,13 @@ static __init int kprobe_trace_self_tests_init(void) trace_probe_event_call(&tk->tp), file); } - ret = trace_run_command("-:testprobe", create_or_delete_trace_kprobe); + ret = create_or_delete_trace_kprobe("-:testprobe"); if (WARN_ON_ONCE(ret)) { pr_warn("error on deleting a probe.\n"); warn++; } - ret = trace_run_command("-:testprobe2", create_or_delete_trace_kprobe); + ret = create_or_delete_trace_kprobe("-:testprobe2"); if (WARN_ON_ONCE(ret)) { pr_warn("error on deleting a probe.\n"); warn++; diff --git a/kernel/trace/trace_mmiotrace.c b/kernel/trace/trace_mmiotrace.c index 84582bf1ed5f..64e77b513697 100644 --- a/kernel/trace/trace_mmiotrace.c +++ b/kernel/trace/trace_mmiotrace.c @@ -5,8 +5,6 @@ * Copyright (C) 2008 Pekka Paalanen <pq@iki.fi> */ -#define DEBUG 1 - #include <linux/kernel.h> #include <linux/mmiotrace.h> #include <linux/pci.h> @@ -300,10 +298,11 @@ static void __trace_mmiotrace_rw(struct trace_array *tr, struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct trace_mmiotrace_rw *entry; - int pc = preempt_count(); + unsigned int trace_ctx; + trace_ctx = tracing_gen_ctx_flags(0); event = trace_buffer_lock_reserve(buffer, TRACE_MMIO_RW, - sizeof(*entry), 0, pc); + sizeof(*entry), trace_ctx); if (!event) { atomic_inc(&dropped_count); return; @@ -312,7 +311,7 @@ static void __trace_mmiotrace_rw(struct trace_array *tr, entry->rw = *rw; if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, 0, pc); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } void mmio_trace_rw(struct mmiotrace_rw *rw) @@ -330,10 +329,11 @@ static void __trace_mmiotrace_map(struct trace_array *tr, struct trace_buffer *buffer = tr->array_buffer.buffer; struct ring_buffer_event *event; struct trace_mmiotrace_map *entry; - int pc = preempt_count(); + unsigned int trace_ctx; + trace_ctx = tracing_gen_ctx_flags(0); event = trace_buffer_lock_reserve(buffer, TRACE_MMIO_MAP, - sizeof(*entry), 0, pc); + sizeof(*entry), trace_ctx); if (!event) { atomic_inc(&dropped_count); return; @@ -342,7 +342,7 @@ static void __trace_mmiotrace_map(struct trace_array *tr, entry->map = *map; if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, 0, pc); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } void mmio_trace_mapping(struct mmiotrace_map *map) diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c index 92b1575ae0ca..61255bad7e01 100644 --- a/kernel/trace/trace_output.c +++ b/kernel/trace/trace_output.c @@ -312,13 +312,23 @@ int trace_raw_output_prep(struct trace_iterator *iter, } EXPORT_SYMBOL(trace_raw_output_prep); +void trace_event_printf(struct trace_iterator *iter, const char *fmt, ...) +{ + va_list ap; + + va_start(ap, fmt); + trace_seq_vprintf(&iter->seq, trace_event_format(iter, fmt), ap); + va_end(ap); +} +EXPORT_SYMBOL(trace_event_printf); + static int trace_output_raw(struct trace_iterator *iter, char *name, char *fmt, va_list ap) { struct trace_seq *s = &iter->seq; trace_seq_printf(s, "%s: ", name); - trace_seq_vprintf(s, fmt, ap); + trace_seq_vprintf(s, trace_event_format(iter, fmt), ap); return trace_handle_return(s); } diff --git a/kernel/trace/trace_probe.c b/kernel/trace/trace_probe.c index d2867ccc6aca..ec589a4612df 100644 --- a/kernel/trace/trace_probe.c +++ b/kernel/trace/trace_probe.c @@ -1134,3 +1134,20 @@ bool trace_probe_match_command_args(struct trace_probe *tp, } return true; } + +int trace_probe_create(const char *raw_command, int (*createfn)(int, const char **)) +{ + int argc = 0, ret = 0; + char **argv; + + argv = argv_split(GFP_KERNEL, raw_command, &argc); + if (!argv) + return -ENOMEM; + + if (argc) + ret = createfn(argc, (const char **)argv); + + argv_free(argv); + + return ret; +} diff --git a/kernel/trace/trace_probe.h b/kernel/trace/trace_probe.h index 2f703a20c724..7ce4027089ee 100644 --- a/kernel/trace/trace_probe.h +++ b/kernel/trace/trace_probe.h @@ -341,6 +341,7 @@ struct event_file_link *trace_probe_get_file_link(struct trace_probe *tp, int trace_probe_compare_arg_type(struct trace_probe *a, struct trace_probe *b); bool trace_probe_match_command_args(struct trace_probe *tp, int argc, const char **argv); +int trace_probe_create(const char *raw_command, int (*createfn)(int, const char **)); #define trace_probe_for_each_link(pos, tp) \ list_for_each_entry(pos, &(tp)->event->files, list) diff --git a/kernel/trace/trace_sched_wakeup.c b/kernel/trace/trace_sched_wakeup.c index c0181066dbe9..e5778d1d7a5b 100644 --- a/kernel/trace/trace_sched_wakeup.c +++ b/kernel/trace/trace_sched_wakeup.c @@ -67,7 +67,7 @@ static bool function_enabled; static int func_prolog_preempt_disable(struct trace_array *tr, struct trace_array_cpu **data, - int *pc) + unsigned int *trace_ctx) { long disabled; int cpu; @@ -75,7 +75,7 @@ func_prolog_preempt_disable(struct trace_array *tr, if (likely(!wakeup_task)) return 0; - *pc = preempt_count(); + *trace_ctx = tracing_gen_ctx(); preempt_disable_notrace(); cpu = raw_smp_processor_id(); @@ -116,8 +116,8 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace) { struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; - unsigned long flags; - int pc, ret = 0; + unsigned int trace_ctx; + int ret = 0; if (ftrace_graph_ignore_func(trace)) return 0; @@ -131,11 +131,10 @@ static int wakeup_graph_entry(struct ftrace_graph_ent *trace) if (ftrace_graph_notrace_addr(trace->func)) return 1; - if (!func_prolog_preempt_disable(tr, &data, &pc)) + if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return 0; - local_save_flags(flags); - ret = __trace_graph_entry(tr, trace, flags, pc); + ret = __trace_graph_entry(tr, trace, trace_ctx); atomic_dec(&data->disabled); preempt_enable_notrace(); @@ -146,16 +145,14 @@ static void wakeup_graph_return(struct ftrace_graph_ret *trace) { struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; - unsigned long flags; - int pc; + unsigned int trace_ctx; ftrace_graph_addr_finish(trace); - if (!func_prolog_preempt_disable(tr, &data, &pc)) + if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return; - local_save_flags(flags); - __trace_graph_return(tr, trace, flags, pc); + __trace_graph_return(tr, trace, trace_ctx); atomic_dec(&data->disabled); preempt_enable_notrace(); @@ -217,13 +214,13 @@ wakeup_tracer_call(unsigned long ip, unsigned long parent_ip, struct trace_array *tr = wakeup_trace; struct trace_array_cpu *data; unsigned long flags; - int pc; + unsigned int trace_ctx; - if (!func_prolog_preempt_disable(tr, &data, &pc)) + if (!func_prolog_preempt_disable(tr, &data, &trace_ctx)) return; local_irq_save(flags); - trace_function(tr, ip, parent_ip, flags, pc); + trace_function(tr, ip, parent_ip, trace_ctx); local_irq_restore(flags); atomic_dec(&data->disabled); @@ -303,12 +300,12 @@ static void wakeup_print_header(struct seq_file *s) static void __trace_function(struct trace_array *tr, unsigned long ip, unsigned long parent_ip, - unsigned long flags, int pc) + unsigned int trace_ctx) { if (is_graph(tr)) - trace_graph_function(tr, ip, parent_ip, flags, pc); + trace_graph_function(tr, ip, parent_ip, trace_ctx); else - trace_function(tr, ip, parent_ip, flags, pc); + trace_function(tr, ip, parent_ip, trace_ctx); } static int wakeup_flag_changed(struct trace_array *tr, u32 mask, int set) @@ -375,7 +372,7 @@ static void tracing_sched_switch_trace(struct trace_array *tr, struct task_struct *prev, struct task_struct *next, - unsigned long flags, int pc) + unsigned int trace_ctx) { struct trace_event_call *call = &event_context_switch; struct trace_buffer *buffer = tr->array_buffer.buffer; @@ -383,7 +380,7 @@ tracing_sched_switch_trace(struct trace_array *tr, struct ctx_switch_entry *entry; event = trace_buffer_lock_reserve(buffer, TRACE_CTX, - sizeof(*entry), flags, pc); + sizeof(*entry), trace_ctx); if (!event) return; entry = ring_buffer_event_data(event); @@ -396,14 +393,14 @@ tracing_sched_switch_trace(struct trace_array *tr, entry->next_cpu = task_cpu(next); if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, flags, pc); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } static void tracing_sched_wakeup_trace(struct trace_array *tr, struct task_struct *wakee, struct task_struct *curr, - unsigned long flags, int pc) + unsigned int trace_ctx) { struct trace_event_call *call = &event_wakeup; struct ring_buffer_event *event; @@ -411,7 +408,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr, struct trace_buffer *buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, TRACE_WAKE, - sizeof(*entry), flags, pc); + sizeof(*entry), trace_ctx); if (!event) return; entry = ring_buffer_event_data(event); @@ -424,7 +421,7 @@ tracing_sched_wakeup_trace(struct trace_array *tr, entry->next_cpu = task_cpu(wakee); if (!call_filter_check_discard(call, entry, buffer, event)) - trace_buffer_unlock_commit(tr, buffer, event, flags, pc); + trace_buffer_unlock_commit(tr, buffer, event, trace_ctx); } static void notrace @@ -436,7 +433,7 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, unsigned long flags; long disabled; int cpu; - int pc; + unsigned int trace_ctx; tracing_record_cmdline(prev); @@ -455,8 +452,6 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, if (next != wakeup_task) return; - pc = preempt_count(); - /* disable local data, not wakeup_cpu data */ cpu = raw_smp_processor_id(); disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); @@ -464,6 +459,8 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, goto out; local_irq_save(flags); + trace_ctx = tracing_gen_ctx_flags(flags); + arch_spin_lock(&wakeup_lock); /* We could race with grabbing wakeup_lock */ @@ -473,9 +470,9 @@ probe_wakeup_sched_switch(void *ignore, bool preempt, /* The task we are waiting for is waking up */ data = per_cpu_ptr(wakeup_trace->array_buffer.data, wakeup_cpu); - __trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, flags, pc); - tracing_sched_switch_trace(wakeup_trace, prev, next, flags, pc); - __trace_stack(wakeup_trace, flags, 0, pc); + __trace_function(wakeup_trace, CALLER_ADDR0, CALLER_ADDR1, trace_ctx); + tracing_sched_switch_trace(wakeup_trace, prev, next, trace_ctx); + __trace_stack(wakeup_trace, trace_ctx, 0); T0 = data->preempt_timestamp; T1 = ftrace_now(cpu); @@ -527,9 +524,8 @@ probe_wakeup(void *ignore, struct task_struct *p) { struct trace_array_cpu *data; int cpu = smp_processor_id(); - unsigned long flags; long disabled; - int pc; + unsigned int trace_ctx; if (likely(!tracer_enabled)) return; @@ -550,11 +546,12 @@ probe_wakeup(void *ignore, struct task_struct *p) (!dl_task(p) && (p->prio >= wakeup_prio || p->prio >= current->prio))) return; - pc = preempt_count(); disabled = atomic_inc_return(&per_cpu_ptr(wakeup_trace->array_buffer.data, cpu)->disabled); if (unlikely(disabled != 1)) goto out; + trace_ctx = tracing_gen_ctx(); + /* interrupts should be off from try_to_wake_up */ arch_spin_lock(&wakeup_lock); @@ -581,19 +578,17 @@ probe_wakeup(void *ignore, struct task_struct *p) wakeup_task = get_task_struct(p); - local_save_flags(flags); - data = per_cpu_ptr(wakeup_trace->array_buffer.data, wakeup_cpu); data->preempt_timestamp = ftrace_now(cpu); - tracing_sched_wakeup_trace(wakeup_trace, p, current, flags, pc); - __trace_stack(wakeup_trace, flags, 0, pc); + tracing_sched_wakeup_trace(wakeup_trace, p, current, trace_ctx); + __trace_stack(wakeup_trace, trace_ctx, 0); /* * We must be careful in using CALLER_ADDR2. But since wake_up * is not called by an assembly function (where as schedule is) * it should be safe to use it here. */ - __trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, flags, pc); + __trace_function(wakeup_trace, CALLER_ADDR1, CALLER_ADDR2, trace_ctx); out_locked: arch_spin_unlock(&wakeup_lock); diff --git a/kernel/trace/trace_syscalls.c b/kernel/trace/trace_syscalls.c index d85a2f0f316b..8bfcd3b09422 100644 --- a/kernel/trace/trace_syscalls.c +++ b/kernel/trace/trace_syscalls.c @@ -298,9 +298,8 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) struct syscall_metadata *sys_data; struct ring_buffer_event *event; struct trace_buffer *buffer; - unsigned long irq_flags; + unsigned int trace_ctx; unsigned long args[6]; - int pc; int syscall_nr; int size; @@ -322,12 +321,11 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) size = sizeof(*entry) + sizeof(unsigned long) * sys_data->nb_args; - local_save_flags(irq_flags); - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, - sys_data->enter_event->event.type, size, irq_flags, pc); + sys_data->enter_event->event.type, size, trace_ctx); if (!event) return; @@ -337,7 +335,7 @@ static void ftrace_syscall_enter(void *data, struct pt_regs *regs, long id) memcpy(entry->args, args, sizeof(unsigned long) * sys_data->nb_args); event_trigger_unlock_commit(trace_file, buffer, event, entry, - irq_flags, pc); + trace_ctx); } static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) @@ -348,8 +346,7 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) struct syscall_metadata *sys_data; struct ring_buffer_event *event; struct trace_buffer *buffer; - unsigned long irq_flags; - int pc; + unsigned int trace_ctx; int syscall_nr; syscall_nr = trace_get_syscall_nr(current, regs); @@ -368,13 +365,12 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) if (!sys_data) return; - local_save_flags(irq_flags); - pc = preempt_count(); + trace_ctx = tracing_gen_ctx(); buffer = tr->array_buffer.buffer; event = trace_buffer_lock_reserve(buffer, sys_data->exit_event->event.type, sizeof(*entry), - irq_flags, pc); + trace_ctx); if (!event) return; @@ -383,7 +379,7 @@ static void ftrace_syscall_exit(void *data, struct pt_regs *regs, long ret) entry->ret = syscall_get_return_value(current, regs); event_trigger_unlock_commit(trace_file, buffer, event, entry, - irq_flags, pc); + trace_ctx); } static int reg_event_syscall_enter(struct trace_event_file *file, diff --git a/kernel/trace/trace_uprobe.c b/kernel/trace/trace_uprobe.c index 3cf7128e1ad3..9b50869a5ddb 100644 --- a/kernel/trace/trace_uprobe.c +++ b/kernel/trace/trace_uprobe.c @@ -34,7 +34,7 @@ struct uprobe_trace_entry_head { #define DATAOF_TRACE_ENTRY(entry, is_return) \ ((void*)(entry) + SIZEOF_TRACE_ENTRY(is_return)) -static int trace_uprobe_create(int argc, const char **argv); +static int trace_uprobe_create(const char *raw_command); static int trace_uprobe_show(struct seq_file *m, struct dyn_event *ev); static int trace_uprobe_release(struct dyn_event *ev); static bool trace_uprobe_is_busy(struct dyn_event *ev); @@ -530,7 +530,7 @@ end: * Argument syntax: * - Add uprobe: p|r[:[GRP/]EVENT] PATH:OFFSET[%return][(REF)] [FETCHARGS] */ -static int trace_uprobe_create(int argc, const char **argv) +static int __trace_uprobe_create(int argc, const char **argv) { struct trace_uprobe *tu; const char *event = NULL, *group = UPROBE_EVENT_SYSTEM; @@ -716,14 +716,19 @@ fail_address_parse: return ret; } -static int create_or_delete_trace_uprobe(int argc, char **argv) +int trace_uprobe_create(const char *raw_command) +{ + return trace_probe_create(raw_command, __trace_uprobe_create); +} + +static int create_or_delete_trace_uprobe(const char *raw_command) { int ret; - if (argv[0][0] == '-') - return dyn_event_release(argc, argv, &trace_uprobe_ops); + if (raw_command[0] == '-') + return dyn_event_release(raw_command, &trace_uprobe_ops); - ret = trace_uprobe_create(argc, (const char **)argv); + ret = trace_uprobe_create(raw_command); return ret == -ECANCELED ? -EINVAL : ret; } @@ -961,7 +966,7 @@ static void __uprobe_trace_func(struct trace_uprobe *tu, esize = SIZEOF_TRACE_ENTRY(is_ret_probe(tu)); size = esize + tu->tp.size + dsize; event = trace_event_buffer_lock_reserve(&buffer, trace_file, - call->event.type, size, 0, 0); + call->event.type, size, 0); if (!event) return; @@ -977,7 +982,7 @@ static void __uprobe_trace_func(struct trace_uprobe *tu, memcpy(data, ucb->buf, tu->tp.size + dsize); - event_trigger_unlock_commit(trace_file, buffer, event, entry, 0, 0); + event_trigger_unlock_commit(trace_file, buffer, event, entry, 0); } /* uprobe handler */ @@ -1635,7 +1640,7 @@ void destroy_local_trace_uprobe(struct trace_event_call *event_call) } #endif /* CONFIG_PERF_EVENTS */ -/* Make a trace interface for controling probe points */ +/* Make a trace interface for controlling probe points */ static __init int init_uprobe_trace(void) { int ret; diff --git a/kernel/tracepoint.c b/kernel/tracepoint.c index 7261fa0f5e3c..9f478d29b926 100644 --- a/kernel/tracepoint.c +++ b/kernel/tracepoint.c @@ -53,6 +53,12 @@ struct tp_probes { struct tracepoint_func probes[]; }; +/* Called in removal of a func but failed to allocate a new tp_funcs */ +static void tp_stub_func(void) +{ + return; +} + static inline void *allocate_probes(int count) { struct tp_probes *p = kmalloc(struct_size(p, probes, count), @@ -130,8 +136,9 @@ func_add(struct tracepoint_func **funcs, struct tracepoint_func *tp_func, int prio) { struct tracepoint_func *old, *new; - int nr_probes = 0; - int pos = -1; + int iter_probes; /* Iterate over old probe array. */ + int nr_probes = 0; /* Counter for probes */ + int pos = -1; /* Insertion position into new array */ if (WARN_ON(!tp_func->func)) return ERR_PTR(-EINVAL); @@ -140,13 +147,13 @@ func_add(struct tracepoint_func **funcs, struct tracepoint_func *tp_func, old = *funcs; if (old) { /* (N -> N+1), (N != 0, 1) probes */ - for (nr_probes = 0; old[nr_probes].func; nr_probes++) { - /* Insert before probes of lower priority */ - if (pos < 0 && old[nr_probes].prio < prio) - pos = nr_probes; - if (old[nr_probes].func == tp_func->func && - old[nr_probes].data == tp_func->data) + for (iter_probes = 0; old[iter_probes].func; iter_probes++) { + if (old[iter_probes].func == tp_stub_func) + continue; /* Skip stub functions. */ + if (old[iter_probes].func == tp_func->func && + old[iter_probes].data == tp_func->data) return ERR_PTR(-EEXIST); + nr_probes++; } } /* + 2 : one for new probe, one for NULL func */ @@ -154,20 +161,24 @@ func_add(struct tracepoint_func **funcs, struct tracepoint_func *tp_func, if (new == NULL) return ERR_PTR(-ENOMEM); if (old) { - if (pos < 0) { - pos = nr_probes; - memcpy(new, old, nr_probes * sizeof(struct tracepoint_func)); - } else { - /* Copy higher priority probes ahead of the new probe */ - memcpy(new, old, pos * sizeof(struct tracepoint_func)); - /* Copy the rest after it. */ - memcpy(new + pos + 1, old + pos, - (nr_probes - pos) * sizeof(struct tracepoint_func)); + nr_probes = 0; + for (iter_probes = 0; old[iter_probes].func; iter_probes++) { + if (old[iter_probes].func == tp_stub_func) + continue; + /* Insert before probes of lower priority */ + if (pos < 0 && old[iter_probes].prio < prio) + pos = nr_probes++; + new[nr_probes++] = old[iter_probes]; } - } else + if (pos < 0) + pos = nr_probes++; + /* nr_probes now points to the end of the new array */ + } else { pos = 0; + nr_probes = 1; /* must point at end of array */ + } new[pos] = *tp_func; - new[nr_probes + 1].func = NULL; + new[nr_probes].func = NULL; *funcs = new; debug_print_probes(*funcs); return old; @@ -188,8 +199,9 @@ static void *func_remove(struct tracepoint_func **funcs, /* (N -> M), (N > 1, M >= 0) probes */ if (tp_func->func) { for (nr_probes = 0; old[nr_probes].func; nr_probes++) { - if (old[nr_probes].func == tp_func->func && - old[nr_probes].data == tp_func->data) + if ((old[nr_probes].func == tp_func->func && + old[nr_probes].data == tp_func->data) || + old[nr_probes].func == tp_stub_func) nr_del++; } } @@ -208,14 +220,27 @@ static void *func_remove(struct tracepoint_func **funcs, /* N -> M, (N > 1, M > 0) */ /* + 1 for NULL */ new = allocate_probes(nr_probes - nr_del + 1); - if (new == NULL) - return ERR_PTR(-ENOMEM); - for (i = 0; old[i].func; i++) - if (old[i].func != tp_func->func - || old[i].data != tp_func->data) - new[j++] = old[i]; - new[nr_probes - nr_del].func = NULL; - *funcs = new; + if (new) { + for (i = 0; old[i].func; i++) { + if ((old[i].func != tp_func->func || + old[i].data != tp_func->data) && + old[i].func != tp_stub_func) + new[j++] = old[i]; + } + new[nr_probes - nr_del].func = NULL; + *funcs = new; + } else { + /* + * Failed to allocate, replace the old function + * with calls to tp_stub_func. + */ + for (i = 0; old[i].func; i++) { + if (old[i].func == tp_func->func && + old[i].data == tp_func->data) + WRITE_ONCE(old[i].func, tp_stub_func); + } + *funcs = old; + } } debug_print_probes(*funcs); return old; @@ -295,10 +320,12 @@ static int tracepoint_remove_func(struct tracepoint *tp, tp_funcs = rcu_dereference_protected(tp->funcs, lockdep_is_held(&tracepoints_mutex)); old = func_remove(&tp_funcs, func); - if (IS_ERR(old)) { - WARN_ON_ONCE(PTR_ERR(old) != -ENOMEM); + if (WARN_ON_ONCE(IS_ERR(old))) return PTR_ERR(old); - } + + if (tp_funcs == old) + /* Failed allocating new tp_funcs, replaced func with stub */ + return 0; if (!tp_funcs) { /* Removed last function */ diff --git a/kernel/watch_queue.c b/kernel/watch_queue.c index 0ef8f65bd2d7..9c9eb20dd2c5 100644 --- a/kernel/watch_queue.c +++ b/kernel/watch_queue.c @@ -413,7 +413,7 @@ static void put_watch(struct watch *watch) } /** - * init_watch_queue - Initialise a watch + * init_watch - Initialise a watch * @watch: The watch to initialise. * @wqueue: The queue to assign. * diff --git a/kernel/workqueue.c b/kernel/workqueue.c index 894bb885b40b..0d150da252e8 100644 --- a/kernel/workqueue.c +++ b/kernel/workqueue.c @@ -2964,8 +2964,8 @@ reflush: if (++flush_cnt == 10 || (flush_cnt % 100 == 0 && flush_cnt <= 1000)) - pr_warn("workqueue %s: drain_workqueue() isn't complete after %u tries\n", - wq->name, flush_cnt); + pr_warn("workqueue %s: %s() isn't complete after %u tries\n", + wq->name, __func__, flush_cnt); mutex_unlock(&wq->mutex); goto reflush; |