Merge branch 'akpm' (patches from Andrew)

Merge updates from Andrew Morton:

 - fsnotify updates

 - ocfs2 updates

 - all of MM

* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (127 commits)
  console: don't prefer first registered if DT specifies stdout-path
  cred: simpler, 1D supplementary groups
  CREDITS: update Pavel's information, add GPG key, remove snail mail address
  mailmap: add Johan Hovold
  .gitattributes: set git diff driver for C source code files
  uprobes: remove function declarations from arch/{mips,s390}
  spelling.txt: "modeled" is spelt correctly
  nmi_backtrace: generate one-line reports for idle cpus
  arch/tile: adopt the new nmi_backtrace framework
  nmi_backtrace: do a local dump_stack() instead of a self-NMI
  nmi_backtrace: add more trigger_*_cpu_backtrace() methods
  min/max: remove sparse warnings when they're nested
  Documentation/filesystems/proc.txt: add more description for maps/smaps
  mm, proc: fix region lost in /proc/self/smaps
  proc: fix timerslack_ns CAP_SYS_NICE check when adjusting self
  proc: add LSM hook checks to /proc/<tid>/timerslack_ns
  proc: relax /proc/<tid>/timerslack_ns capability requirements
  meminfo: break apart a very long seq_printf with #ifdefs
  seq/proc: modify seq_put_decimal_[u]ll to take a const char *, not char
  proc: faster /proc/*/status
  ...

5 files changed, 361 insertions, 62 deletions

diff --git a/tools/testing/selftests/vm/.gitignore b/tools/testing/selftests/vm/.gitignore
index a937a9d26b60..142c565bb351 100644
--- a/tools/testing/selftests/vm/.gitignore
+++ b/tools/testing/selftests/vm/.gitignore
@@ -7,3 +7,4 @@ mlock2-tests
 on-fault-limit
 transhuge-stress
 userfaultfd
+mlock-intersect-test
diff --git a/tools/testing/selftests/vm/Makefile b/tools/testing/selftests/vm/Makefile
index e4bb1de1d526..bbab7f4664ac 100644
--- a/tools/testing/selftests/vm/Makefile
+++ b/tools/testing/selftests/vm/Makefile
@@ -10,6 +10,7 @@ BINARIES += on-fault-limit
 BINARIES += thuge-gen
 BINARIES += transhuge-stress
 BINARIES += userfaultfd
+BINARIES += mlock-random-test
 
 all: $(BINARIES)
 %: %.c
@@ -17,6 +18,9 @@ all: $(BINARIES)
 userfaultfd: userfaultfd.c ../../../../usr/include/linux/kernel.h
 	$(CC) $(CFLAGS) -O2 -o $@ $< -lpthread
 
+mlock-random-test: mlock-random-test.c
+	$(CC) $(CFLAGS) -o $@ $< -lcap
+
 ../../../../usr/include/linux/kernel.h:
 	make -C ../../../.. headers_install
 
diff --git a/tools/testing/selftests/vm/mlock-random-test.c b/tools/testing/selftests/vm/mlock-random-test.c
new file mode 100644
index 000000000000..83de4f58d262
--- /dev/null
+++ b/tools/testing/selftests/vm/mlock-random-test.c
@@ -0,0 +1,293 @@
+/*
+ * It tests the mlock/mlock2() when they are invoked
+ * on randomly memory region.
+ */
+#include <unistd.h>
+#include <sys/resource.h>
+#include <sys/capability.h>
+#include <sys/mman.h>
+#include <fcntl.h>
+#include <string.h>
+#include <sys/ipc.h>
+#include <sys/shm.h>
+#include <time.h>
+#include "mlock2.h"
+
+#define CHUNK_UNIT (128 * 1024)
+#define MLOCK_RLIMIT_SIZE (CHUNK_UNIT * 2)
+#define MLOCK_WITHIN_LIMIT_SIZE CHUNK_UNIT
+#define MLOCK_OUTOF_LIMIT_SIZE (CHUNK_UNIT * 3)
+
+#define TEST_LOOP 100
+#define PAGE_ALIGN(size, ps) (((size) + ((ps) - 1)) & ~((ps) - 1))
+
+int set_cap_limits(rlim_t max)
+{
+	struct rlimit new;
+	cap_t cap = cap_init();
+
+	new.rlim_cur = max;
+	new.rlim_max = max;
+	if (setrlimit(RLIMIT_MEMLOCK, &new)) {
+		perror("setrlimit() returns error\n");
+		return -1;
+	}
+
+	/* drop capabilities including CAP_IPC_LOCK */
+	if (cap_set_proc(cap)) {
+		perror("cap_set_proc() returns error\n");
+		return -2;
+	}
+
+	return 0;
+}
+
+int get_proc_locked_vm_size(void)
+{
+	FILE *f;
+	int ret = -1;
+	char line[1024] = {0};
+	unsigned long lock_size = 0;
+
+	f = fopen("/proc/self/status", "r");
+	if (!f) {
+		perror("fopen");
+		return -1;
+	}
+
+	while (fgets(line, 1024, f)) {
+		if (strstr(line, "VmLck")) {
+			ret = sscanf(line, "VmLck:\t%8lu kB", &lock_size);
+			if (ret <= 0) {
+				printf("sscanf() on VmLck error: %s: %d\n",
+						line, ret);
+				fclose(f);
+				return -1;
+			}
+			fclose(f);
+			return (int)(lock_size << 10);
+		}
+	}
+
+	perror("cann't parse VmLck in /proc/self/status\n");
+	fclose(f);
+	return -1;
+}
+
+/*
+ * Get the MMUPageSize of the memory region including input
+ * address from proc file.
+ *
+ * return value: on error case, 0 will be returned.
+ * Otherwise the page size(in bytes) is returned.
+ */
+int get_proc_page_size(unsigned long addr)
+{
+	FILE *smaps;
+	char *line;
+	unsigned long mmupage_size = 0;
+	size_t size;
+
+	smaps = seek_to_smaps_entry(addr);
+	if (!smaps) {
+		printf("Unable to parse /proc/self/smaps\n");
+		return 0;
+	}
+
+	while (getline(&line, &size, smaps) > 0) {
+		if (!strstr(line, "MMUPageSize")) {
+			free(line);
+			line = NULL;
+			size = 0;
+			continue;
+		}
+
+		/* found the MMUPageSize of this section */
+		if (sscanf(line, "MMUPageSize:    %8lu kB",
+					&mmupage_size) < 1) {
+			printf("Unable to parse smaps entry for Size:%s\n",
+					line);
+			break;
+		}
+
+	}
+	free(line);
+	if (smaps)
+		fclose(smaps);
+	return mmupage_size << 10;
+}
+
+/*
+ * Test mlock/mlock2() on provided memory chunk.
+ * It expects the mlock/mlock2() to be successful (within rlimit)
+ *
+ * With allocated memory chunk [p, p + alloc_size), this
+ * test will choose start/len randomly to perform mlock/mlock2
+ * [start, start +  len] memory range. The range is within range
+ * of the allocated chunk.
+ *
+ * The memory region size alloc_size is within the rlimit.
+ * So we always expect a success of mlock/mlock2.
+ *
+ * VmLck is assumed to be 0 before this test.
+ *
+ *    return value: 0 - success
+ *    else: failure
+ */
+int test_mlock_within_limit(char *p, int alloc_size)
+{
+	int i;
+	int ret = 0;
+	int locked_vm_size = 0;
+	struct rlimit cur;
+	int page_size = 0;
+
+	getrlimit(RLIMIT_MEMLOCK, &cur);
+	if (cur.rlim_cur < alloc_size) {
+		printf("alloc_size[%d] < %u rlimit,lead to mlock failure\n",
+				alloc_size, (unsigned int)cur.rlim_cur);
+		return -1;
+	}
+
+	srand(time(NULL));
+	for (i = 0; i < TEST_LOOP; i++) {
+		/*
+		 * - choose mlock/mlock2 randomly
+		 * - choose lock_size randomly but lock_size < alloc_size
+		 * - choose start_offset randomly but p+start_offset+lock_size
+		 *   < p+alloc_size
+		 */
+		int is_mlock = !!(rand() % 2);
+		int lock_size = rand() % alloc_size;
+		int start_offset = rand() % (alloc_size - lock_size);
+
+		if (is_mlock)
+			ret = mlock(p + start_offset, lock_size);
+		else
+			ret = mlock2_(p + start_offset, lock_size,
+				       MLOCK_ONFAULT);
+
+		if (ret) {
+			printf("%s() failure at |%p(%d)| mlock:|%p(%d)|\n",
+					is_mlock ? "mlock" : "mlock2",
+					p, alloc_size,
+					p + start_offset, lock_size);
+			return ret;
+		}
+	}
+
+	/*
+	 * Check VmLck left by the tests.
+	 */
+	locked_vm_size = get_proc_locked_vm_size();
+	page_size = get_proc_page_size((unsigned long)p);
+	if (page_size == 0) {
+		printf("cannot get proc MMUPageSize\n");
+		return -1;
+	}
+
+	if (locked_vm_size > PAGE_ALIGN(alloc_size, page_size) + page_size) {
+		printf("test_mlock_within_limit() left VmLck:%d on %d chunk\n",
+				locked_vm_size, alloc_size);
+		return -1;
+	}
+
+	return 0;
+}
+
+
+/*
+ * We expect the mlock/mlock2() to be fail (outof limitation)
+ *
+ * With allocated memory chunk [p, p + alloc_size), this
+ * test will randomly choose start/len and perform mlock/mlock2
+ * on [start, start+len] range.
+ *
+ * The memory region size alloc_size is above the rlimit.
+ * And the len to be locked is higher than rlimit.
+ * So we always expect a failure of mlock/mlock2.
+ * No locked page number should be increased as a side effect.
+ *
+ *    return value: 0 - success
+ *    else: failure
+ */
+int test_mlock_outof_limit(char *p, int alloc_size)
+{
+	int i;
+	int ret = 0;
+	int locked_vm_size = 0, old_locked_vm_size = 0;
+	struct rlimit cur;
+
+	getrlimit(RLIMIT_MEMLOCK, &cur);
+	if (cur.rlim_cur >= alloc_size) {
+		printf("alloc_size[%d] >%u rlimit, violates test condition\n",
+				alloc_size, (unsigned int)cur.rlim_cur);
+		return -1;
+	}
+
+	old_locked_vm_size = get_proc_locked_vm_size();
+	srand(time(NULL));
+	for (i = 0; i < TEST_LOOP; i++) {
+		int is_mlock = !!(rand() % 2);
+		int lock_size = (rand() % (alloc_size - cur.rlim_cur))
+			+ cur.rlim_cur;
+		int start_offset = rand() % (alloc_size - lock_size);
+
+		if (is_mlock)
+			ret = mlock(p + start_offset, lock_size);
+		else
+			ret = mlock2_(p + start_offset, lock_size,
+					MLOCK_ONFAULT);
+		if (ret == 0) {
+			printf("%s() succeeds? on %p(%d) mlock%p(%d)\n",
+					is_mlock ? "mlock" : "mlock2",
+					p, alloc_size,
+					p + start_offset, lock_size);
+			return -1;
+		}
+	}
+
+	locked_vm_size = get_proc_locked_vm_size();
+	if (locked_vm_size != old_locked_vm_size) {
+		printf("tests leads to new mlocked page: old[%d], new[%d]\n",
+				old_locked_vm_size,
+				locked_vm_size);
+		return -1;
+	}
+
+	return 0;
+}
+
+int main(int argc, char **argv)
+{
+	char *p = NULL;
+	int ret = 0;
+
+	if (set_cap_limits(MLOCK_RLIMIT_SIZE))
+		return -1;
+
+	p = malloc(MLOCK_WITHIN_LIMIT_SIZE);
+	if (p == NULL) {
+		perror("malloc() failure\n");
+		return -1;
+	}
+	ret = test_mlock_within_limit(p, MLOCK_WITHIN_LIMIT_SIZE);
+	if (ret)
+		return ret;
+	munlock(p, MLOCK_WITHIN_LIMIT_SIZE);
+	free(p);
+
+
+	p = malloc(MLOCK_OUTOF_LIMIT_SIZE);
+	if (p == NULL) {
+		perror("malloc() failure\n");
+		return -1;
+	}
+	ret = test_mlock_outof_limit(p, MLOCK_OUTOF_LIMIT_SIZE);
+	if (ret)
+		return ret;
+	munlock(p, MLOCK_OUTOF_LIMIT_SIZE);
+	free(p);
+
+	return 0;
+}
diff --git a/tools/testing/selftests/vm/mlock2-tests.c b/tools/testing/selftests/vm/mlock2-tests.c
index 02ca5e0177c5..ff0cda2b19c9 100644
--- a/tools/testing/selftests/vm/mlock2-tests.c
+++ b/tools/testing/selftests/vm/mlock2-tests.c
@@ -1,33 +1,12 @@
 #define _GNU_SOURCE
 #include <sys/mman.h>
 #include <stdint.h>
-#include <stdio.h>
-#include <stdlib.h>
 #include <unistd.h>
 #include <string.h>
 #include <sys/time.h>
 #include <sys/resource.h>
-#include <syscall.h>
-#include <errno.h>
 #include <stdbool.h>
-
-#ifndef MLOCK_ONFAULT
-#define MLOCK_ONFAULT 1
-#endif
-
-#ifndef MCL_ONFAULT
-#define MCL_ONFAULT (MCL_FUTURE << 1)
-#endif
-
-static int mlock2_(void *start, size_t len, int flags)
-{
-#ifdef __NR_mlock2
-	return syscall(__NR_mlock2, start, len, flags);
-#else
-	errno = ENOSYS;
-	return -1;
-#endif
-}
+#include "mlock2.h"
 
 struct vm_boundaries {
 	unsigned long start;
@@ -138,46 +117,6 @@ static uint64_t get_kpageflags(unsigned long pfn)
 	return flags;
 }
 
-static FILE *seek_to_smaps_entry(unsigned long addr)
-{
-	FILE *file;
-	char *line = NULL;
-	size_t size = 0;
-	unsigned long start, end;
-	char perms[5];
-	unsigned long offset;
-	char dev[32];
-	unsigned long inode;
-	char path[BUFSIZ];
-
-	file = fopen("/proc/self/smaps", "r");
-	if (!file) {
-		perror("fopen smaps");
-		_exit(1);
-	}
-
-	while (getline(&line, &size, file) > 0) {
-		if (sscanf(line, "%lx-%lx %s %lx %s %lu %s\n",
-			   &start, &end, perms, &offset, dev, &inode, path) < 6)
-			goto next;
-
-		if (start <= addr && addr < end)
-			goto out;
-
-next:
-		free(line);
-		line = NULL;
-		size = 0;
-	}
-
-	fclose(file);
-	file = NULL;
-
-out:
-	free(line);
-	return file;
-}
-
 #define VMFLAGS "VmFlags:"
 
 static bool is_vmflag_set(unsigned long addr, const char *vmflag)
diff --git a/tools/testing/selftests/vm/mlock2.h b/tools/testing/selftests/vm/mlock2.h
new file mode 100644
index 000000000000..7ee062929d3e
--- /dev/null
+++ b/tools/testing/selftests/vm/mlock2.h
@@ -0,0 +1,62 @@
+#include <syscall.h>
+#include <errno.h>
+#include <stdio.h>
+#include <stdlib.h>
+
+#ifndef MLOCK_ONFAULT
+#define MLOCK_ONFAULT 1
+#endif
+
+#ifndef MCL_ONFAULT
+#define MCL_ONFAULT (MCL_FUTURE << 1)
+#endif
+
+static int mlock2_(void *start, size_t len, int flags)
+{
+#ifdef __NR_mlock2
+	return syscall(__NR_mlock2, start, len, flags);
+#else
+	errno = ENOSYS;
+	return -1;
+#endif
+}
+
+static FILE *seek_to_smaps_entry(unsigned long addr)
+{
+	FILE *file;
+	char *line = NULL;
+	size_t size = 0;
+	unsigned long start, end;
+	char perms[5];
+	unsigned long offset;
+	char dev[32];
+	unsigned long inode;
+	char path[BUFSIZ];
+
+	file = fopen("/proc/self/smaps", "r");
+	if (!file) {
+		perror("fopen smaps");
+		_exit(1);
+	}
+
+	while (getline(&line, &size, file) > 0) {
+		if (sscanf(line, "%lx-%lx %s %lx %s %lu %s\n",
+			   &start, &end, perms, &offset, dev, &inode, path) < 6)
+			goto next;
+
+		if (start <= addr && addr < end)
+			goto out;
+
+next:
+		free(line);
+		line = NULL;
+		size = 0;
+	}
+
+	fclose(file);
+	file = NULL;
+
+out:
+	free(line);
+	return file;
+}