Merge tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-tracegrafted

Pull tracing fixes from Steven Rostedt:

 - Don't destroy user event fields when removal fails

   User event fields are destroyed before the event is removed from
   visibility. But that can fail leaving the still visible event with no
   fields. Move the destroying of the fields to after the event is
   successfully removed from visibility.

 - Initialize function graph state is fork before calling
   copy_exec_state()

   For non-CLONE_VM forks, copy_exec_state() allocates a new
   task_exec_state. If that allocation fails, ftrace_graph_exit_task()
   will free the tasks ret_stack pointer. Since that pointer is still
   using the parent's ret_stack, it mistakenly frees the parent's
   pointer too.

   Call ftrace_graph_init() on the task first which will NULL out the
   new tasks's ret_stack and if the copy fails, it will not free
   anything.

 - Remove FGRAPH_MAX_INDEX

   The macro FGRAPH_MAX_INDEX was added but never used. Remove it.

 - Save ent_size in function graph printing of nested functions

   The function graph tracer needs to look at the next event to see if
   the next event is the return of the current function entry. If it is,
   it prints a single line:

	ktime_get();

   Otherwise it prints it like a nested function:

	tick_nohz_irq_exit() {
	    ktime_get();
	    kcpustat_irq_exit();
	}

   In order to look at the next event, it must save the current event so
   that it has the information to print from it. It saves the event in
   the iterator descriptor called "ent". What it doesn't save is the
   ent_size of the event which is now used to know if the function graph
   arguments are to be printed. The peek doesn't save the size so the
   size used happens to be that of the size of the last event that was
   seen.

   Save the entry event size in the iterator descriptor so that the
   correct size is used.

 - Fix several errors with freeing data in the histogram code

   The histogram code had a lot of leaked or or incorrect accounting
   when failures happen. Correct them.

 - Fix histogram regression of .percent and .graph modifiers

   Up until 6.3 histogram values could have "percent" or "graph"
   modifiers that changed how they were printed. But a change that added
   restricting histograms values from being strings, stack traces and
   other modifiers inadvertently prevented them from using the percent
   and graph modifiers, which were legal use cases for values.

   Put back the percent and graph modifiers.

 - Fix various typos in the comments

 - Set the trace_clock before initializing a histogram with clock
   argument

   The histogram API allows the user to specific which trace clock to
   use via a "clock=" string. The histogram is set up first before the
   clock is checked. If the passed in clock is not valid, it exits
   without fully fixing up the histogram leaving it on the list and a
   use-after-free can trigger.

   Update the clock argument first and if it fails then exit gracefully
   before the histogram trigger is placed on any lists.

 - Restore :mod: trailer after parsing in ftrace_set_clr_event

   The function ftrace_set_clr_event() modifies the parse string and
   needs to put it back to what was passed in. It searches for ":mod:"
   via a strsep() but fails to put back the first ':' in the string.

   Add back the ':' in the passed in string.

 - Take trace_array reference when opening a tracer options file

   The options files are dynamically created and some tracers add their
   own options. When a tracer adds their own list of options, the
   trace_array holding them has an array to hold the list of options for
   each tracer. This array increases in size via a krealloc(), and the
   new entry gets a newly allocated array to hold the options of the new
   tracer being added.

   The element in each entry of the tracer's option array holds a
   pointer back to the trace_array, a pointer to the tracer it is
   associated to, a pointer to the flags of the option.

   The issue is that these arrays are freed when the trace_array is
   freed when its instance it represents is removed from the instances
   directory. There's a race that an open of one of these options files
   can happen when the instance is being removed.

   Add a new helper function to be called by the open function of the
   options file to iterate all existing trace_arrays under a lock and
   find the one that has the given option element in one of it's tracer
   arrays. If found, then update the associated trace_array's reference
   counter to keep it from being freed. If not found, have the open call
   return -ENODEV.

 - Disable interrupts when acquiring the lock in rb_wake_up_waiters()

   The function rb_wake_up_waiters() assumes it will be called in
   interrupt context and does not disable irqs when taking
   cpu_buffer->reader_lock, which can be called in hard interrupt
   context. The issue is in PREEMPT_RT, this function is called in
   thread context leaving this lock open to a deadlock.

   Take the lock with interrupts disabled.

 - Use rcu_assign_pointer() for tmp_ops filter hash

   The tmp_ops used in update_ftrace_direct_mod() assigns its
   filter_hash field directly, but that field is annotated as __rcu and
   sparse complains. Assign it with rcu_assign_pointer()

 - Fix use-after-free in enable_trigger_private_data_free()

   The trace_event_call is accessed through the event_trigger_data's
   trace_event_file pointer to put the trace_event_call on freeing. The
   issue is that the trace_event_file data may have been freed already
   causing a use-after-free. Add a field to the event_trigger_data that
   points directly to the trace_event_call so that it can decrement its
   reference directly without needing to go through the
   trace_event_file.

 - Fix accounting of buffer data remote headers

   trace_buffer_desc_size() and trace_remote_alloc_buffer() undercount
   the number of pages is needed for the asked for size as it doesn't
   take into account the meta data on each page. Add a helper function
   to do the calculation properly and use that in these functions.

 - Catch nr_page_va overflow in ring_buffer_desc sizing

   The number of pages per remote ring buffer is capped by
   ring_buffer_desc::nr_page_va (32 bits). A buffer_size large enough to
   overflow that field would silently allocate a descriptor smaller than
   what was asked for.

 - Do not resize the subbuf order if any per_cpu buffer is disabled

   The mmapping of ring buffers disables resizing the subbuffers, but it
   is done per-cpu whereas the subbuf size change is done for all the
   per_cpu buffers under the buffer->mutex. It could change the size of
   some while the mapping is happening on others. Have the resize of the
   subbuf order check all the per_cpu buffers under the lock to see if
   any of them is disabled before starting and causing an inconsistency
   between buffers that are being mapped.

* tag 'trace-v7.3-rc2' of git://git.kernel.org/pub/scm/linux/kernel/git/trace/linux-trace: (25 commits)
  ring-buffer: Check resize_disabled before publishing the new subbuf order
  tracing/remotes: Catch nr_page_va overflow in ring_buffer_desc sizing
  tracing/remotes: Account for ring buffer page header in size calculation
  tracing: Don't dereference trace_event_file in deferred trigger free
  ftrace: Use rcu_assign_pointer() for tmp_ops filter hash
  ring-buffer: Acquire the lock with irqsave in rb_wake_up_waiters()
  tracing: Take trace_array reference when opening a tracer options file
  tracing: Fix ring_buffer_read_page_size() kernel-doc
  tracing: Restore :mod: trailer after parsing in ftrace_set_clr_event()
  tracing: Fix memory corruption from a "STACKTRACE" histogram key
  tracing: Fix memory corruption from the histogram stacktrace modifier
  tracing: Undo the registration when enabling the histogram trigger fails
  tracing: Take the reference before publishing the named histogram trigger
  tracing: Set the trace clock before registering the histogram trigger
  tracing: Fix typo "preceeded" in comment
  tracing: Fix typo "availabe" in comment
  tracing: Let histogram values keep the percent and graph modifiers
  tracing: Keep the entry count when the histogram stats allocation fails
  tracing: Free histogram the field rejected for a bad modifier
  tracing: Free histogram the var ref when its initialization fails
  ...

1 files changed, 440 insertions, 0 deletions

diff --git a/scripts/basic/fixdep.c b/scripts/basic/fixdep.c
new file mode 100644
index 000000000..54063d980
--- /dev/null
+++ b/scripts/basic/fixdep.c
@@ -0,0 +1,440 @@
+/*
+ * "Optimize" a list of dependencies as spit out by gcc -MD
+ * for the kernel build
+ * ===========================================================================
+ *
+ * Author       Kai Germaschewski
+ * Copyright    2002 by Kai Germaschewski  <kai.germaschewski@gmx.de>
+ *
+ * This software may be used and distributed according to the terms
+ * of the GNU General Public License, incorporated herein by reference.
+ *
+ *
+ * Introduction:
+ *
+ * gcc produces a very nice and correct list of dependencies which
+ * tells make when to remake a file.
+ *
+ * However, to use this list as-is has the drawback that virtually
+ * every file in the kernel includes autoconf.h.
+ *
+ * If the user re-runs make *config, autoconf.h will be
+ * regenerated.  make notices that and will rebuild every file which
+ * includes autoconf.h, i.e. basically all files. This is extremely
+ * annoying if the user just changed CONFIG_USER_DRIVER from n to m.
+ *
+ * So we play the same trick that "mkdep" played before. We replace
+ * the dependency on autoconf.h by a dependency on every config
+ * option which is mentioned in any of the listed prerequisites.
+ *
+ * kconfig populates a tree in include/config/ with an empty file
+ * for each config symbol and when the configuration is updated
+ * the files representing changed config options are touched
+ * which then let make pick up the changes and the files that use
+ * the config symbols are rebuilt.
+ *
+ * So if the user changes their CONFIG_USER_DRIVER option, only the objects
+ * which depend on "include/config/USER_DRIVER" will be rebuilt,
+ * so most likely only the user's driver ;-)
+ *
+ * The idea above dates, by the way, back to Michael E Chastain, AFAIK.
+ *
+ * So to get dependencies right, there are two issues:
+ * o if any of the files the compiler read changed, we need to rebuild
+ * o if the command line given to the compile the file changed, we
+ *   better rebuild as well.
+ *
+ * The former is handled by using the -MD output, the later by saving
+ * the command line used to compile the old object and comparing it
+ * to the one we would now use.
+ *
+ * Again, also this idea is pretty old and has been discussed on
+ * kbuild-devel a long time ago. I don't have a sensibly working
+ * internet connection right now, so I rather don't mention names
+ * without double checking.
+ *
+ * This code here has been based partially based on mkdep.c, which
+ * says the following about its history:
+ *
+ *   Copyright abandoned, Michael Chastain, <mailto:mec@shout.net>.
+ *   This is a C version of syncdep.pl by Werner Almesberger.
+ *
+ *
+ * It is invoked as
+ *
+ *   fixdep <depfile> <target> <cmdline>
+ *
+ * and will read the dependency file <depfile>
+ *
+ * The transformed dependency snipped is written to stdout.
+ *
+ * It first generates a line
+ *
+ *   savedcmd_<target> = <cmdline>
+ *
+ * and then basically copies the .<target>.d file to stdout, in the
+ * process filtering out the dependency on autoconf.h and adding
+ * dependencies on include/config/MY_OPTION for every
+ * CONFIG_MY_OPTION encountered in any of the prerequisites.
+ *
+ * We don't even try to really parse the header files, but
+ * merely grep, i.e. if CONFIG_FOO is mentioned in a comment, it will
+ * be picked up as well. It's not a problem with respect to
+ * correctness, since that can only give too many dependencies, thus
+ * we cannot miss a rebuild. Since people tend to not mention totally
+ * unrelated CONFIG_ options all over the place, it's not an
+ * efficiency problem either.
+ *
+ * (Note: it'd be easy to port over the complete mkdep state machine,
+ *  but I don't think the added complexity is worth it)
+ */
+
+#include <sys/types.h>
+#include <sys/stat.h>
+#include <unistd.h>
+#include <fcntl.h>
+#include <string.h>
+#include <stdbool.h>
+#include <stdlib.h>
+#include <stdio.h>
+#include <ctype.h>
+
+#include <xalloc.h>
+
+static void usage(void)
+{
+	fprintf(stderr, "Usage: fixdep <depfile> <target> <cmdline>\n");
+	exit(1);
+}
+
+struct item {
+	struct item	*next;
+	unsigned int	len;
+	unsigned int	hash;
+	char		name[];
+};
+
+#define HASHSZ 256
+static struct item *config_hashtab[HASHSZ], *file_hashtab[HASHSZ];
+
+static unsigned int strhash(const char *str, unsigned int sz)
+{
+	/* fnv32 hash */
+	unsigned int i, hash = 2166136261U;
+
+	for (i = 0; i < sz; i++)
+		hash = (hash ^ str[i]) * 0x01000193;
+	return hash;
+}
+
+/*
+ * Add a new value to the configuration string.
+ */
+static void add_to_hashtable(const char *name, int len, unsigned int hash,
+			     struct item *hashtab[])
+{
+	struct item *aux;
+
+	aux = xmalloc(sizeof(*aux) + len);
+	memcpy(aux->name, name, len);
+	aux->len = len;
+	aux->hash = hash;
+	aux->next = hashtab[hash % HASHSZ];
+	hashtab[hash % HASHSZ] = aux;
+}
+
+/*
+ * Lookup a string in the hash table. If found, just return true.
+ * If not, add it to the hashtable and return false.
+ */
+static bool in_hashtable(const char *name, int len, struct item *hashtab[])
+{
+	struct item *aux;
+	unsigned int hash = strhash(name, len);
+
+	for (aux = hashtab[hash % HASHSZ]; aux; aux = aux->next) {
+		if (aux->hash == hash && aux->len == len &&
+		    memcmp(aux->name, name, len) == 0)
+			return true;
+	}
+
+	add_to_hashtable(name, len, hash, hashtab);
+
+	return false;
+}
+
+/*
+ * Record the use of a CONFIG_* word.
+ */
+static void use_config(const char *m, int slen)
+{
+	if (in_hashtable(m, slen, config_hashtab))
+		return;
+
+	/* Print out a dependency path from a symbol name. */
+	printf("    $(wildcard include/config/%.*s) \\\n", slen, m);
+}
+
+/* test if s ends in sub */
+static int str_ends_with(const char *s, int slen, const char *sub)
+{
+	int sublen = strlen(sub);
+
+	if (sublen > slen)
+		return 0;
+
+	return !memcmp(s + slen - sublen, sub, sublen);
+}
+
+static void parse_config_file(const char *p)
+{
+	const char *q, *r;
+	const char *start = p;
+
+	while ((p = strstr(p, "CONFIG_"))) {
+		if (p > start && (isalnum(p[-1]) || p[-1] == '_')) {
+			p += 7;
+			continue;
+		}
+		p += 7;
+		q = p;
+		while (isalnum(*q) || *q == '_')
+			q++;
+		if (str_ends_with(p, q - p, "_MODULE"))
+			r = q - 7;
+		else
+			r = q;
+		if (r > p)
+			use_config(p, r - p);
+		p = q;
+	}
+}
+
+static void *read_file(const char *filename)
+{
+	struct stat st;
+	int fd;
+	char *buf;
+
+	fd = open(filename, O_RDONLY);
+	if (fd < 0) {
+		fprintf(stderr, "fixdep: error opening file: ");
+		perror(filename);
+		exit(2);
+	}
+	if (fstat(fd, &st) < 0) {
+		fprintf(stderr, "fixdep: error fstat'ing file: ");
+		perror(filename);
+		exit(2);
+	}
+	buf = xmalloc(st.st_size + 1);
+	if (read(fd, buf, st.st_size) != st.st_size) {
+		perror("fixdep: read");
+		exit(2);
+	}
+	buf[st.st_size] = '\0';
+	close(fd);
+
+	return buf;
+}
+
+/* Ignore certain dependencies */
+static int is_ignored_file(const char *s, int len)
+{
+	return str_ends_with(s, len, "include/generated/autoconf.h");
+}
+
+/* Do not parse these files */
+static int is_no_parse_file(const char *s, int len)
+{
+	/* rustc may list binary files in dep-info */
+	return str_ends_with(s, len, ".rlib") ||
+	       str_ends_with(s, len, ".rmeta") ||
+	       str_ends_with(s, len, ".so");
+}
+
+/*
+ * Important: The below generated source_foo.o and deps_foo.o variable
+ * assignments are parsed not only by make, but also by the rather simple
+ * parser in scripts/mod/sumversion.c.
+ */
+static void parse_dep_file(char *p, const char *target)
+{
+	bool saw_any_target = false;
+	bool is_target = true;
+	bool is_source = false;
+	bool need_parse;
+	char *q, saved_c;
+
+	while (*p) {
+		/* handle some special characters first. */
+		switch (*p) {
+		case '#':
+			/*
+			 * skip comments.
+			 * rustc may emit comments to dep-info.
+			 */
+			p++;
+			while (*p != '\0' && *p != '\n') {
+				/*
+				 * escaped newlines continue the comment across
+				 * multiple lines.
+				 */
+				if (*p == '\\')
+					p++;
+				p++;
+			}
+			continue;
+		case ' ':
+		case '\t':
+			/* skip whitespaces */
+			p++;
+			continue;
+		case '\\':
+			/*
+			 * backslash/newline combinations continue the
+			 * statement. Skip it just like a whitespace.
+			 */
+			if (*(p + 1) == '\n') {
+				p += 2;
+				continue;
+			}
+			break;
+		case '\n':
+			/*
+			 * Makefiles use a line-based syntax, where the newline
+			 * is the end of a statement. After seeing a newline,
+			 * we expect the next token is a target.
+			 */
+			p++;
+			is_target = true;
+			continue;
+		case ':':
+			/*
+			 * assume the first dependency after a colon as the
+			 * source file.
+			 */
+			p++;
+			is_target = false;
+			is_source = true;
+			continue;
+		}
+
+		/* find the end of the token */
+		q = p;
+		while (*q != ' ' && *q != '\t' && *q != '\n' && *q != '#' && *q != ':') {
+			if (*q == '\\') {
+				/*
+				 * backslash/newline combinations work like as
+				 * a whitespace, so this is the end of token.
+				 */
+				if (*(q + 1) == '\n')
+					break;
+
+				/* escaped special characters */
+				if (*(q + 1) == '#' || *(q + 1) == ':') {
+					memmove(p + 1, p, q - p);
+					p++;
+				}
+
+				q++;
+			}
+
+			if (*q == '\0')
+				break;
+			q++;
+		}
+
+		/* Just discard the target */
+		if (is_target) {
+			p = q;
+			continue;
+		}
+
+		saved_c = *q;
+		*q = '\0';
+		need_parse = false;
+
+		/*
+		 * Do not list the source file as dependency, so that kbuild is
+		 * not confused if a .c file is rewritten into .S or vice versa.
+		 * Storing it in source_* is needed for modpost to compute
+		 * srcversions.
+		 */
+		if (is_source) {
+			/*
+			 * The DT build rule concatenates multiple dep files.
+			 * When processing them, only process the first source
+			 * name, which will be the original one, and ignore any
+			 * other source names, which will be intermediate
+			 * temporary files.
+			 *
+			 * rustc emits the same dependency list for each
+			 * emission type. It is enough to list the source name
+			 * just once.
+			 */
+			if (!saw_any_target) {
+				saw_any_target = true;
+				printf("source_%s := %s\n\n", target, p);
+				printf("deps_%s := \\\n", target);
+				need_parse = true;
+			}
+		} else if (!is_ignored_file(p, q - p) &&
+			   !in_hashtable(p, q - p, file_hashtab)) {
+			printf("  %s \\\n", p);
+			need_parse = true;
+		}
+
+		if (need_parse && !is_no_parse_file(p, q - p)) {
+			void *buf;
+
+			buf = read_file(p);
+			parse_config_file(buf);
+			free(buf);
+		}
+
+		is_source = false;
+		*q = saved_c;
+		p = q;
+	}
+
+	if (!saw_any_target) {
+		fprintf(stderr, "fixdep: parse error; no targets found\n");
+		exit(1);
+	}
+
+	printf("\n%s: $(deps_%s)\n\n", target, target);
+	printf("$(deps_%s):\n", target);
+}
+
+int main(int argc, char *argv[])
+{
+	const char *depfile, *target, *cmdline;
+	void *buf;
+
+	if (argc != 4)
+		usage();
+
+	depfile = argv[1];
+	target = argv[2];
+	cmdline = argv[3];
+
+	printf("savedcmd_%s := %s\n\n", target, cmdline);
+
+	buf = read_file(depfile);
+	parse_dep_file(buf, target);
+	free(buf);
+
+	fflush(stdout);
+
+	/*
+	 * In the intended usage, the stdout is redirected to .*.cmd files.
+	 * Call ferror() to catch errors such as "No space left on device".
+	 */
+	if (ferror(stdout)) {
+		fprintf(stderr, "fixdep: not all data was written to the output\n");
+		exit(1);
+	}
+
+	return 0;
+}