ftrace: test for running of recordmcount.pl twice on an object

Impact: fix failure of dynamic function tracer selftest In a course of development, a developer does several makes on their kernel. Sometimes, the make might do something abnormal. In the case of running the recordmcount.pl script on an object twice, the script will duplicate all the calls to mcount in the __mcount_loc section. On boot up, the dynamic function tracer is careful when it modifies code, and performs several consistency checks. One is to not modify the call site if it is not what it expects it to be. If a function call site is listed twice, the first entry will convert the site to a nop, and the second will fail because it expected to see a call to mcount, but instead it sees a nop. Thus, the function tracer is disabled. Eric Sesterhenn reported seeing: [ 1.055440] ftrace: converting mcount calls to 0f 1f 44 00 00 [ 1.055568] ftrace: allocating 29418 entries in 116 pages [ 1.061000] ------------[ cut here ]------------ [ 1.061000] WARNING: at kernel/trace/ftrace.c:441 [...] [ 1.060000] ---[ end trace 4eaa2a86a8e2da23 ]--- [ 1.060000] ftrace failed to modify [<c0118072>] check_corruption+0x3/0x2d [ 1.060000] actual: 0f:1f:44:00:00 This warning shows that check_corruption+0x3 already had a nop in its place (0x0f1f440000). After compiling another kernel the problem went away. Later Eric Paris notice the same type of issue. Luckily, he saved the vmlinux file that caused it. In the file we found a bunch of duplicate mcount call site records, which lead us to the script. Perhaps this problem only happens to people named Eric. This patch changes the script to test if the __mcount_loc already exists in the object file, and if it does, it will print out an error message and kill the compile. Reported-by: Eric Sesterhenn <snakebyte@gmx.de> Reported-by: Eric Paris <eparis@redhat.com> Signed-off-by: Steven Rostedt <srostedt@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>

ftrace: test for running of recordmcount.pl twice on an object
Impact: fix failure of dynamic function tracer selftest In a course of development, a developer does several makes on their kernel. Sometimes, the make might do something abnormal. In the case of running the recordmcount.pl script on an object twice, the script will duplicate all the calls to mcount in the __mcount_loc section. On boot up, the dynamic function tracer is careful when it modifies code, and performs several consistency checks. One is to not modify the call site if it is not what it expects it to be. If a function call site is listed twice, the first entry will convert the site to a nop, and the second will fail because it expected to see a call to mcount, but instead it sees a nop. Thus, the function tracer is disabled. Eric Sesterhenn reported seeing: [ 1.055440] ftrace: converting mcount calls to 0f 1f 44 00 00 [ 1.055568] ftrace: allocating 29418 entries in 116 pages [ 1.061000] ------------[ cut here ]------------ [ 1.061000] WARNING: at kernel/trace/ftrace.c:441 [...] [ 1.060000] ---[ end trace 4eaa2a86a8e2da23 ]--- [ 1.060000] ftrace failed to modify [<c0118072>] check_corruption+0x3/0x2d [ 1.060000] actual: 0f:1f:44:00:00 This warning shows that check_corruption+0x3 already had a nop in its place (0x0f1f440000). After compiling another kernel the problem went away. Later Eric Paris notice the same type of issue. Luckily, he saved the vmlinux file that caused it. In the file we found a bunch of duplicate mcount call site records, which lead us to the script. Perhaps this problem only happens to people named Eric. This patch changes the script to test if the __mcount_loc already exists in the object file, and if it does, it will print out an error message and kill the compile. Reported-by: Eric Sesterhenn <snakebyte@gmx.de> Reported-by: Eric Paris <eparis@redhat.com> Signed-off-by: Steven Rostedt <srostedt@redhat.com> Signed-off-by: Ingo Molnar <mingo@elte.hu>
Steven Rostedt · Ingo Molnar
1 parent 4092762aeb
Showing 1 changed file with 18 additions and 3 deletions Side-by-side Diff
scripts/recordmcount.pl
@@ -101,7 +101,7 @@
 my $V = '0.1';
  
 if ($#ARGV < 7) {
-	print "usage: $P arch objdump objcopy cc ld nm rm mv is_module inputfile\n";
+	print "usage: $P arch bits objdump objcopy cc ld nm rm mv is_module inputfile\n";
 	print "version: $V\n";
 	exit(1);
 }
@@ -275,7 +275,6 @@
 	"\tDisabling local function references.\n";
 }
  
-
 #
 # Step 1: find all the local (static functions) and weak symbols.
 #        't' is local, 'w/W' is weak (we never use a weak function)
  
  
@@ -343,13 +342,16 @@
 #
 # Step 2: find the sections and mcount call sites
 #
-open(IN, "$objdump -dr $inputfile|") || die "error running $objdump";
+open(IN, "$objdump -hdr $inputfile|") || die "error running $objdump";
  
 my $text;
  
+my $read_headers = 1;
+
 while (<IN>) {
     # is it a section?
     if (/$section_regex/) {
+	$read_headers = 0;
  
 	# Only record text sections that we know are safe
 	if (defined($text_sections{$1})) {
@@ -383,6 +385,19 @@
 		$ref_func = $text;
 	    }
 	}
+    } elsif ($read_headers && /$mcount_section/) {
+	#
+	# Somehow the make process can execute this script on an
+	# object twice. If it does, we would duplicate the mcount
+	# section and it will cause the function tracer self test
+	# to fail. Check if the mcount section exists, and if it does,
+	# warn and exit.
+	#
+	print STDERR "ERROR: $mcount_section already in $inputfile\n" .
+	    "\tThis may be an indication that your build is corrupted.\n" .
+	    "\tDelete $inputfile and try again. If the same object file\n" .
+	    "\tstill causes an issue, then disable CONFIG_DYNAMIC_FTRACE.\n";
+	exit(-1);
     }
  
     # is this a call site to mcount? If so, record it to print later
...	...	@@ -101,7 +101,7 @@
101	101	my $V = '0.1';
102	102
103	103	if ($#ARGV < 7) {
104		- print "usage: $P arch objdump objcopy cc ld nm rm mv is_module inputfile\n";
	104	+ print "usage: $P arch bits objdump objcopy cc ld nm rm mv is_module inputfile\n";
105	105	print "version: $V\n";
106	106	exit(1);
107	107	}
...	...	@@ -275,7 +275,6 @@
275	275	"\tDisabling local function references.\n";
276	276	}
277	277
278		-
279	278	#
280	279	# Step 1: find all the local (static functions) and weak symbols.
281	280	# 't' is local, 'w/W' is weak (we never use a weak function)
282	281
283	282
...	...	@@ -343,13 +342,16 @@
343	342	#
344	343	# Step 2: find the sections and mcount call sites
345	344	#
346		-open(IN, "$objdump -dr $inputfile\|") \|\| die "error running $objdump";
	345	+open(IN, "$objdump -hdr $inputfile\|") \|\| die "error running $objdump";
347	346
348	347	my $text;
349	348
	349	+my $read_headers = 1;
	350	+
350	351	while (<IN>) {
351	352	# is it a section?
352	353	if (/$section_regex/) {
	354	+ $read_headers = 0;
353	355
354	356	# Only record text sections that we know are safe
355	357	if (defined($text_sections{$1})) {
...	...	@@ -383,6 +385,19 @@
383	385	$ref_func = $text;
384	386	}
385	387	}
	388	+ } elsif ($read_headers && /$mcount_section/) {
	389	+ #
	390	+ # Somehow the make process can execute this script on an
	391	+ # object twice. If it does, we would duplicate the mcount
	392	+ # section and it will cause the function tracer self test
	393	+ # to fail. Check if the mcount section exists, and if it does,
	394	+ # warn and exit.
	395	+ #
	396	+ print STDERR "ERROR: $mcount_section already in $inputfile\n" .
	397	+ "\tThis may be an indication that your build is corrupted.\n" .
	398	+ "\tDelete $inputfile and try again. If the same object file\n" .
	399	+ "\tstill causes an issue, then disable CONFIG_DYNAMIC_FTRACE.\n";
	400	+ exit(-1);
386	401	}
387	402
388	403	# is this a call site to mcount? If so, record it to print later