Merge tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Fix error output of boot instance creation failure

   Currently if a boot instance creation fails, instead of printing out
   the name of the instance that failed, it prints "(null)". That is
   because it prints "cur_str" that had already been processed by
   strsep(). Print the saved name instead.

   While at it, print the error code of the failure.

 - Fix use-after-free for same named historgrams

   Histograms can be named so that they can be used in multiple events.
   But if the named histogram has a variable attached, the second event
   that uses the named histogram which duplicates it and needs to free
   the original after duplication leaves the old variable in place and
   still visible. If another histogram uses than variable, it will use
   the stale one which will try to reference the freed duplicate
   histogram and crash the kernel.

   Free the duplicate variables along with the duplicated histogram
   data.

 - Check return value of kthread_run() in event self test

   The events self tests uses a kthread for testing but does not check
   if it succeeded in creating a kthread. If the kthread creation were
   to fail, the code will still try to call kthread_stop() on the error
   returned.

 - Fix race between reading trace_pipe and updating subbuffer size

   If a user is reading the trace_pipe file at the same time they update
   the ring buffer sub-buffer size, can cause the trace_pipe read to
   read stale data. Add trace_access_lock() around updating the ring
   buffer sub-buffer size.

 - Fix eventfs_inode on failure path in creation of the events directory

   In the creation of the "events" directory, if after allocating the
   eventfs_inode a failure is detected, it calls cleanup_ei() which
   calls free_ei(). The free_ei() will test if eventfs_inode being freed
   has no children. It is a bug if it does. But on the failure case of
   the creation of the "events" directory, the children lists have not
   yet been initialized and the free will trigger a warning because
   list_empty() on an uninitialized list returns false.

   Move the initialization into init_ei() where it makes more sense and
   makes sure that a created eventfs_inode has its lists initialized
   upon creation.

 - Check return value of kthread_run() in ftrace direct sample code

   The sample code that shows how to use the ftrace direct calls does
   not test the return of kthread_run() to see if it succeeds. Return a
   failure if the kthread_run() doesn't succeed.

 - Clear user events state on fork in case of alloc failure

   On fork, the child gets a pointer to the parent's user events state.
   It makes a copy of it then updates the child's pointer to it. But if
   the allocation fails, the duplication function leaves the child with
   a pointer to its parent's descriptor. When the child cleans up its
   data, it will free the parent's descriptor while the parent is still
   using it.

   In the duplication function, set the child's user_event_mm to NULL
   before testing if the allocation succeeded, and when it exits it will
   not free the parent's descriptor.

 - Fix retry exhaustion in simple ring buffer reader swap

   simple_ring_buffer_swap_reader_page() starts with retry set to 8 and
   post-decrements it only after a failed link replacement. On the final
   attempt, a successful replacement leaves retry at zero, while a
   failed replacement leaves it at -1.

   But the check for success expects the retry value to be non-zero and
   exits with an error on zero. This is the opposite result. Fix it.

 - Fail nicely when the remote swap_reader_page() returns an error

   Currently, if the swap_reader_page() of a remote buffer fails, it
   triggers a WARN_ON_ONCE() and continues normally. Instead, have it
   exit with an error and a pr_warn() print instead of a full WARNING.

* tag 'trace-v7.3-2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace:
  ring-buffer: Stop remote reader update when page swap fails
  tracing: Fix retry exhaustion in simple ring buffer reader swap
  tracing/user_events: Clear copied tracing state before fork duplication
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-multi-modify
  samples/ftrace: Fix kthread_stop() on ERR_PTR in ftrace-direct-modify
  eventfs: Initialize ei->children and ei->list in init_ei()
  tracing: Fix use-after-free in trace_pipe read on sub-buffer order change
  tracing: Fix crash passing ERR_PTR to kthread_stop()
  tracing: Fix use-after-free with same-name named triggers
  tracing: Fix logged instance name on creation failure

1 files changed, 399 insertions, 0 deletions

diff --git a/security/apparmor/task.c b/security/apparmor/task.c
new file mode 100644
index 000000000..e16ff4130
--- /dev/null
+++ b/security/apparmor/task.c
@@ -0,0 +1,399 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/*
+ * AppArmor security module
+ *
+ * This file contains AppArmor task related definitions and mediation
+ *
+ * Copyright 2017 Canonical Ltd.
+ *
+ * TODO
+ * If a task uses change_hat it currently does not return to the old
+ * cred or task context but instead creates a new one.  Ideally the task
+ * should return to the previous cred if it has not been modified.
+ */
+
+#include <linux/gfp.h>
+#include <linux/ptrace.h>
+#include <linux/task_work.h>
+
+#include "include/path.h"
+#include "include/audit.h"
+#include "include/cred.h"
+#include "include/policy.h"
+#include "include/task.h"
+
+/**
+ * aa_get_task_label - Get another task's label
+ * @task: task to query  (NOT NULL)
+ *
+ * Returns: counted reference to @task's label
+ */
+struct aa_label *aa_get_task_label(struct task_struct *task)
+{
+	struct aa_label *p;
+
+	rcu_read_lock();
+	p = aa_get_newest_cred_label(__task_cred(task));
+	rcu_read_unlock();
+
+	return p;
+}
+
+/**
+ * aa_replace_current_label - replace the current tasks label
+ * @label: new label  (NOT NULL)
+ *
+ * Returns: 0 or error on failure
+ */
+int aa_replace_current_label(struct aa_label *label)
+{
+	struct aa_label *old = aa_current_raw_label();
+	struct aa_task_ctx *ctx = task_ctx(current);
+	struct cred *new;
+
+	AA_BUG(!label);
+
+	if (old == label)
+		return 0;
+
+	if (current_cred() != current_real_cred())
+		return -EBUSY;
+
+	new  = prepare_creds();
+	if (!new)
+		return -ENOMEM;
+
+	if (ctx->nnp && label_is_stale(ctx->nnp)) {
+		struct aa_label *tmp = ctx->nnp;
+
+		ctx->nnp = aa_get_newest_label(tmp);
+		aa_put_label(tmp);
+	}
+	if (unconfined(label) || (labels_ns(old) != labels_ns(label)))
+		/*
+		 * if switching to unconfined or a different label namespace
+		 * clear out context state
+		 */
+		aa_clear_task_ctx_trans(task_ctx(current));
+
+	/*
+	 * be careful switching cred label, when racing replacement it
+	 * is possible that the cred labels's->proxy->label is the reference
+	 * keeping @label valid, so make sure to get its reference before
+	 * dropping the reference on the cred's label
+	 */
+	aa_get_label(label);
+	aa_put_label(cred_label(new));
+	set_cred_label(new, label);
+
+	commit_creds(new);
+	return 0;
+}
+
+static void aa_replace_stale_label_tw_func(struct callback_head *tw)
+{
+	struct aa_task_ctx *ctx = task_ctx(current);
+	struct aa_label *label;
+
+	ctx->label_replacement_pending = false;
+	label = aa_current_raw_label();
+	if (!label_is_stale(label))
+		return;
+	label = aa_get_newest_label(label);
+	aa_replace_current_label(label);
+	aa_put_label(label);
+}
+
+/* replace the current task's stale label on syscall return */
+void aa_schedule_stale_label_replacement(void)
+{
+	struct aa_task_ctx *ctx = task_ctx(current);
+
+	if (ctx->label_replacement_pending)
+		return;
+	init_task_work(&ctx->label_replacement_tw, aa_replace_stale_label_tw_func);
+	if (task_work_add(current, &ctx->label_replacement_tw, TWA_RESUME) == 0)
+		ctx->label_replacement_pending = true;
+}
+
+
+/**
+ * aa_set_current_onexec - set the tasks change_profile to happen onexec
+ * @label: system label to set at exec  (MAYBE NULL to clear value)
+ * @stack: whether stacking should be done
+ */
+void aa_set_current_onexec(struct aa_label *label, bool stack)
+{
+	struct aa_task_ctx *ctx = task_ctx(current);
+
+	aa_get_label(label);
+	aa_put_label(ctx->onexec);
+	ctx->onexec = label;
+	ctx->token = stack;
+}
+
+/**
+ * aa_set_current_hat - set the current tasks hat
+ * @label: label to set as the current hat  (NOT NULL)
+ * @token: token value that must be specified to change from the hat
+ *
+ * Do switch of tasks hat.  If the task is currently in a hat
+ * validate the token to match.
+ *
+ * Returns: 0 or error on failure
+ */
+int aa_set_current_hat(struct aa_label *label, u64 token)
+{
+	struct aa_task_ctx *ctx = task_ctx(current);
+	struct cred *new;
+
+	new = prepare_creds();
+	if (!new)
+		return -ENOMEM;
+	AA_BUG(!label);
+
+	if (!ctx->previous) {
+		/* transfer refcount */
+		ctx->previous = cred_label(new);
+		ctx->token = token;
+	} else if (ctx->token == token) {
+		aa_put_label(cred_label(new));
+	} else {
+		/* previous_profile && ctx->token != token */
+		abort_creds(new);
+		return -EACCES;
+	}
+
+	set_cred_label(new, aa_get_newest_label(label));
+	/* clear exec on switching context */
+	aa_put_label(ctx->onexec);
+	ctx->onexec = NULL;
+
+	commit_creds(new);
+	return 0;
+}
+
+/**
+ * aa_restore_previous_label - exit from hat context restoring previous label
+ * @token: the token that must be matched to exit hat context
+ *
+ * Attempt to return out of a hat to the previous label.  The token
+ * must match the stored token value.
+ *
+ * Returns: 0 or error of failure
+ */
+int aa_restore_previous_label(u64 token)
+{
+	struct aa_task_ctx *ctx = task_ctx(current);
+	struct cred *new;
+
+	if (ctx->token != token)
+		return -EACCES;
+	/* ignore restores when there is no saved label */
+	if (!ctx->previous)
+		return 0;
+
+	new = prepare_creds();
+	if (!new)
+		return -ENOMEM;
+
+	aa_put_label(cred_label(new));
+	set_cred_label(new, aa_get_newest_label(ctx->previous));
+	AA_BUG(!cred_label(new));
+	/* clear exec && prev information when restoring to previous context */
+	aa_clear_task_ctx_trans(ctx);
+
+	commit_creds(new);
+
+	return 0;
+}
+
+/**
+ * audit_ptrace_mask - convert mask to permission string
+ * @mask: permission mask to convert
+ *
+ * Returns: pointer to static string
+ */
+static const char *audit_ptrace_mask(u32 mask)
+{
+	switch (mask) {
+	case MAY_READ:
+		return "read";
+	case MAY_WRITE:
+		return "trace";
+	case AA_MAY_BE_READ:
+		return "readby";
+	case AA_MAY_BE_TRACED:
+		return "tracedby";
+	}
+	return "";
+}
+
+/* call back to audit ptrace fields */
+static void audit_ptrace_cb(struct audit_buffer *ab, void *va)
+{
+	struct common_audit_data *sa = va;
+	struct apparmor_audit_data *ad = aad(sa);
+
+	if (ad->request & AA_PTRACE_PERM_MASK) {
+		audit_log_format(ab, " requested_mask=\"%s\"",
+				 audit_ptrace_mask(ad->request));
+
+		if (ad->denied & AA_PTRACE_PERM_MASK) {
+			audit_log_format(ab, " denied_mask=\"%s\"",
+					 audit_ptrace_mask(ad->denied));
+		}
+	}
+	audit_log_format(ab, " peer=");
+	aa_label_xaudit(ab, labels_ns(ad->subj_label), ad->peer,
+			FLAGS_NONE, GFP_ATOMIC);
+}
+
+/* assumes check for RULE_MEDIATES is already done */
+/* TODO: conditionals */
+static int profile_ptrace_perm(const struct cred *cred,
+			       struct aa_profile *profile,
+			       struct aa_label *peer, u32 request,
+			       struct apparmor_audit_data *ad)
+{
+	struct aa_ruleset *rules = profile->label.rules[0];
+	struct aa_perms perms = { };
+
+	ad->subj_cred = cred;
+	ad->peer = peer;
+	aa_profile_match_label(profile, rules, peer, AA_CLASS_PTRACE, request,
+			       &perms);
+	aa_apply_modes_to_perms(profile, &perms);
+	return aa_check_perms(profile, &perms, request, ad, audit_ptrace_cb);
+}
+
+static int profile_tracee_perm(const struct cred *cred,
+			       struct aa_profile *tracee,
+			       struct aa_label *tracer, u32 request,
+			       struct apparmor_audit_data *ad)
+{
+	if (profile_unconfined(tracee) || unconfined(tracer) ||
+	    !label_mediates(&tracee->label, AA_CLASS_PTRACE))
+		return 0;
+
+	return profile_ptrace_perm(cred, tracee, tracer, request, ad);
+}
+
+static int profile_tracer_perm(const struct cred *cred,
+			       struct aa_profile *tracer,
+			       struct aa_label *tracee, u32 request,
+			       struct apparmor_audit_data *ad)
+{
+	if (profile_unconfined(tracer))
+		return 0;
+
+	if (label_mediates(&tracer->label, AA_CLASS_PTRACE))
+		return profile_ptrace_perm(cred, tracer, tracee, request, ad);
+
+	/* profile uses the old style capability check for ptrace */
+	if (&tracer->label == tracee)
+		return 0;
+
+	ad->subj_label = &tracer->label;
+	ad->peer = tracee;
+	ad->request = 0;
+	ad->error = aa_capable(cred, &tracer->label, CAP_SYS_PTRACE,
+			       CAP_OPT_NONE);
+
+	return aa_audit(AUDIT_APPARMOR_AUTO, tracer, ad, audit_ptrace_cb);
+}
+
+/**
+ * aa_may_ptrace - test if tracer task can trace the tracee
+ * @tracer_cred: cred of task doing the tracing  (NOT NULL)
+ * @tracer: label of the task doing the tracing  (NOT NULL)
+ * @tracee_cred: cred of task to be traced
+ * @tracee: task label to be traced
+ * @request: permission request
+ *
+ * Returns: %0 else error code if permission denied or error
+ */
+int aa_may_ptrace(const struct cred *tracer_cred, struct aa_label *tracer,
+		  const struct cred *tracee_cred, struct aa_label *tracee,
+		  u32 request)
+{
+	struct aa_profile *profile;
+	u32 xrequest = request << PTRACE_PERM_SHIFT;
+	DEFINE_AUDIT_DATA(sa, LSM_AUDIT_DATA_NONE, AA_CLASS_PTRACE, OP_PTRACE);
+
+	return xcheck_labels(tracer, tracee, profile,
+			profile_tracer_perm(tracer_cred, profile, tracee,
+					    request, &sa),
+			profile_tracee_perm(tracee_cred, profile, tracer,
+					    xrequest, &sa));
+}
+
+static const char *get_current_exe_path(char *buffer, int buffer_size)
+{
+	struct file *exe_file;
+	struct path p;
+	const char *path_str;
+
+	exe_file = get_task_exe_file(current);
+	if (!exe_file)
+		return ERR_PTR(-ENOENT);
+	p = exe_file->f_path;
+	path_get(&p);
+
+	if (aa_path_name(&p, FLAG_VIEW_SUBNS, buffer, &path_str, NULL, NULL))
+		path_str = ERR_PTR(-ENOMEM);
+
+	fput(exe_file);
+	path_put(&p);
+
+	return path_str;
+}
+
+/* call back to audit ptrace fields */
+static void audit_ns_cb(struct audit_buffer *ab, void *va)
+{
+	struct apparmor_audit_data *ad = aad_of_va(va);
+	char *buffer;
+	const char *path;
+
+	if (ad->request & AA_USERNS_CREATE)
+		audit_log_format(ab, " requested=\"userns_create\"");
+
+	if (ad->denied & AA_USERNS_CREATE)
+		audit_log_format(ab, " denied=\"userns_create\"");
+
+	buffer = aa_get_buffer(false);
+	if (!buffer)
+		return; // OOM
+	path = get_current_exe_path(buffer, aa_g_path_max);
+	if (!IS_ERR(path))
+		audit_log_format(ab, " execpath=\"%s\"", path);
+	aa_put_buffer(buffer);
+}
+
+int aa_profile_ns_perm(struct aa_profile *profile,
+		       struct apparmor_audit_data *ad,
+		       u32 request)
+{
+	struct aa_perms perms = { };
+	int error = 0;
+
+	ad->subj_label = &profile->label;
+	ad->request = request;
+
+	if (!profile_unconfined(profile)) {
+		struct aa_ruleset *rules = profile->label.rules[0];
+		aa_state_t state;
+
+		state = RULE_MEDIATES(rules, ad->class);
+		if (!state)
+			/* TODO: add flag to complain about unmediated */
+			return 0;
+		perms = *aa_lookup_perms(rules->policy, state);
+		aa_apply_modes_to_perms(profile, &perms);
+		error = aa_check_perms(profile, &perms, request, ad,
+				       audit_ns_cb);
+	}
+
+	return error;
+}