Merge tag 'close-range-v5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux

Pull close_range() implementation from Christian Brauner:
 "This adds the close_range() syscall. It allows to efficiently close a
  range of file descriptors up to all file descriptors of a calling
  task.

  This is coordinated with the FreeBSD folks which have copied our
  version of this syscall and in the meantime have already merged it in
  April 2019:

    https://reviews.freebsd.org/D21627
    https://svnweb.freebsd.org/base?view=revision&revision=359836

  The syscall originally came up in a discussion around the new mount
  API and making new file descriptor types cloexec by default. During
  this discussion, Al suggested the close_range() syscall.

  First, it helps to close all file descriptors of an exec()ing task.
  This can be done safely via (quoting Al's example from [1] verbatim):

        /* that exec is sensitive */
        unshare(CLONE_FILES);
        /* we don't want anything past stderr here */
        close_range(3, ~0U);
        execve(....);

  The code snippet above is one way of working around the problem that
  file descriptors are not cloexec by default. This is aggravated by the
  fact that we can't just switch them over without massively regressing
  userspace. For a whole class of programs having an in-kernel method of
  closing all file descriptors is very helpful (e.g. demons, service
  managers, programming language standard libraries, container managers
  etc.).

  Second, it allows userspace to avoid implementing closing all file
  descriptors by parsing through /proc/<pid>/fd/* and calling close() on
  each file descriptor and other hacks. From looking at various
  large(ish) userspace code bases this or similar patterns are very
  common in service managers, container runtimes, and programming
  language runtimes/standard libraries such as Python or Rust.

  In addition, the syscall will also work for tasks that do not have
  procfs mounted and on kernels that do not have procfs support compiled
  in. In such situations the only way to make sure that all file
  descriptors are closed is to call close() on each file descriptor up
  to UINT_MAX or RLIMIT_NOFILE, OPEN_MAX trickery.

  Based on Linus' suggestion close_range() also comes with a new flag
  CLOSE_RANGE_UNSHARE to more elegantly handle file descriptor dropping
  right before exec. This would usually be expressed in the sequence:

        unshare(CLONE_FILES);
        close_range(3, ~0U);

  as pointed out by Linus it might be desirable to have this be a part
  of close_range() itself under a new flag CLOSE_RANGE_UNSHARE which
  gets especially handy when we're closing all file descriptors above a
  certain threshold.

  Test-suite as always included"

* tag 'close-range-v5.9' of git://git.kernel.org/pub/scm/linux/kernel/git/brauner/linux:
  tests: add CLOSE_RANGE_UNSHARE tests
  close_range: add CLOSE_RANGE_UNSHARE
  tests: add close_range() tests
  arch: wire-up close_range()
  open: add close_range()

4 files changed, 236 insertions, 0 deletions

diff --git a/tools/testing/selftests/Makefile b/tools/testing/selftests/Makefile
index 227ca78a5b7f..017ce2a7ae36 100644
--- a/tools/testing/selftests/Makefile
+++ b/tools/testing/selftests/Makefile
@@ -6,6 +6,7 @@ TARGETS += breakpoints
 TARGETS += capabilities
 TARGETS += cgroup
 TARGETS += clone3
+TARGETS += core
 TARGETS += cpufreq
 TARGETS += cpu-hotplug
 TARGETS += drivers/dma-buf
diff --git a/tools/testing/selftests/core/.gitignore b/tools/testing/selftests/core/.gitignore
new file mode 100644
index 000000000000..6e6712ce5817
--- /dev/null
+++ b/tools/testing/selftests/core/.gitignore
@@ -0,0 +1 @@
+close_range_test
diff --git a/tools/testing/selftests/core/Makefile b/tools/testing/selftests/core/Makefile
new file mode 100644
index 000000000000..f6f2d6f473c6
--- /dev/null
+++ b/tools/testing/selftests/core/Makefile
@@ -0,0 +1,7 @@
+# SPDX-License-Identifier: GPL-2.0-only
+CFLAGS += -g -I../../../../usr/include/
+
+TEST_GEN_PROGS := close_range_test
+
+include ../lib.mk
+
diff --git a/tools/testing/selftests/core/close_range_test.c b/tools/testing/selftests/core/close_range_test.c
new file mode 100644
index 000000000000..c99b98b0d461
--- /dev/null
+++ b/tools/testing/selftests/core/close_range_test.c
@@ -0,0 +1,227 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#define _GNU_SOURCE
+#include <errno.h>
+#include <fcntl.h>
+#include <linux/kernel.h>
+#include <limits.h>
+#include <stdbool.h>
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include <syscall.h>
+#include <unistd.h>
+
+#include "../kselftest_harness.h"
+#include "../clone3/clone3_selftests.h"
+
+#ifndef __NR_close_range
+#define __NR_close_range -1
+#endif
+
+#ifndef CLOSE_RANGE_UNSHARE
+#define CLOSE_RANGE_UNSHARE	(1U << 1)
+#endif
+
+static inline int sys_close_range(unsigned int fd, unsigned int max_fd,
+				  unsigned int flags)
+{
+	return syscall(__NR_close_range, fd, max_fd, flags);
+}
+
+#ifndef ARRAY_SIZE
+#define ARRAY_SIZE(x) (sizeof(x) / sizeof((x)[0]))
+#endif
+
+TEST(close_range)
+{
+	int i, ret;
+	int open_fds[101];
+
+	for (i = 0; i < ARRAY_SIZE(open_fds); i++) {
+		int fd;
+
+		fd = open("/dev/null", O_RDONLY | O_CLOEXEC);
+		ASSERT_GE(fd, 0) {
+			if (errno == ENOENT)
+				XFAIL(return, "Skipping test since /dev/null does not exist");
+		}
+
+		open_fds[i] = fd;
+	}
+
+	EXPECT_EQ(-1, sys_close_range(open_fds[0], open_fds[100], -1)) {
+		if (errno == ENOSYS)
+			XFAIL(return, "close_range() syscall not supported");
+	}
+
+	EXPECT_EQ(0, sys_close_range(open_fds[0], open_fds[50], 0));
+
+	for (i = 0; i <= 50; i++)
+		EXPECT_EQ(-1, fcntl(open_fds[i], F_GETFL));
+
+	for (i = 51; i <= 100; i++)
+		EXPECT_GT(fcntl(open_fds[i], F_GETFL), -1);
+
+	/* create a couple of gaps */
+	close(57);
+	close(78);
+	close(81);
+	close(82);
+	close(84);
+	close(90);
+
+	EXPECT_EQ(0, sys_close_range(open_fds[51], open_fds[92], 0));
+
+	for (i = 51; i <= 92; i++)
+		EXPECT_EQ(-1, fcntl(open_fds[i], F_GETFL));
+
+	for (i = 93; i <= 100; i++)
+		EXPECT_GT(fcntl(open_fds[i], F_GETFL), -1);
+
+	/* test that the kernel caps and still closes all fds */
+	EXPECT_EQ(0, sys_close_range(open_fds[93], open_fds[99], 0));
+
+	for (i = 93; i <= 99; i++)
+		EXPECT_EQ(-1, fcntl(open_fds[i], F_GETFL));
+
+	EXPECT_GT(fcntl(open_fds[i], F_GETFL), -1);
+
+	EXPECT_EQ(0, sys_close_range(open_fds[100], open_fds[100], 0));
+
+	EXPECT_EQ(-1, fcntl(open_fds[100], F_GETFL));
+}
+
+TEST(close_range_unshare)
+{
+	int i, ret, status;
+	pid_t pid;
+	int open_fds[101];
+	struct clone_args args = {
+		.flags = CLONE_FILES,
+		.exit_signal = SIGCHLD,
+	};
+
+	for (i = 0; i < ARRAY_SIZE(open_fds); i++) {
+		int fd;
+
+		fd = open("/dev/null", O_RDONLY | O_CLOEXEC);
+		ASSERT_GE(fd, 0) {
+			if (errno == ENOENT)
+				XFAIL(return, "Skipping test since /dev/null does not exist");
+		}
+
+		open_fds[i] = fd;
+	}
+
+	pid = sys_clone3(&args, sizeof(args));
+	ASSERT_GE(pid, 0);
+
+	if (pid == 0) {
+		ret = sys_close_range(open_fds[0], open_fds[50],
+				      CLOSE_RANGE_UNSHARE);
+		if (ret)
+			exit(EXIT_FAILURE);
+
+		for (i = 0; i <= 50; i++)
+			if (fcntl(open_fds[i], F_GETFL) != -1)
+				exit(EXIT_FAILURE);
+
+		for (i = 51; i <= 100; i++)
+			if (fcntl(open_fds[i], F_GETFL) == -1)
+				exit(EXIT_FAILURE);
+
+		/* create a couple of gaps */
+		close(57);
+		close(78);
+		close(81);
+		close(82);
+		close(84);
+		close(90);
+
+		ret = sys_close_range(open_fds[51], open_fds[92],
+				      CLOSE_RANGE_UNSHARE);
+		if (ret)
+			exit(EXIT_FAILURE);
+
+		for (i = 51; i <= 92; i++)
+			if (fcntl(open_fds[i], F_GETFL) != -1)
+				exit(EXIT_FAILURE);
+
+		for (i = 93; i <= 100; i++)
+			if (fcntl(open_fds[i], F_GETFL) == -1)
+				exit(EXIT_FAILURE);
+
+		/* test that the kernel caps and still closes all fds */
+		ret = sys_close_range(open_fds[93], open_fds[99],
+				      CLOSE_RANGE_UNSHARE);
+		if (ret)
+			exit(EXIT_FAILURE);
+
+		for (i = 93; i <= 99; i++)
+			if (fcntl(open_fds[i], F_GETFL) != -1)
+				exit(EXIT_FAILURE);
+
+		if (fcntl(open_fds[100], F_GETFL) == -1)
+			exit(EXIT_FAILURE);
+
+		ret = sys_close_range(open_fds[100], open_fds[100],
+				      CLOSE_RANGE_UNSHARE);
+		if (ret)
+			exit(EXIT_FAILURE);
+
+		if (fcntl(open_fds[100], F_GETFL) != -1)
+			exit(EXIT_FAILURE);
+
+		exit(EXIT_SUCCESS);
+	}
+
+	EXPECT_EQ(waitpid(pid, &status, 0), pid);
+	EXPECT_EQ(true, WIFEXITED(status));
+	EXPECT_EQ(0, WEXITSTATUS(status));
+}
+
+TEST(close_range_unshare_capped)
+{
+	int i, ret, status;
+	pid_t pid;
+	int open_fds[101];
+	struct clone_args args = {
+		.flags = CLONE_FILES,
+		.exit_signal = SIGCHLD,
+	};
+
+	for (i = 0; i < ARRAY_SIZE(open_fds); i++) {
+		int fd;
+
+		fd = open("/dev/null", O_RDONLY | O_CLOEXEC);
+		ASSERT_GE(fd, 0) {
+			if (errno == ENOENT)
+				XFAIL(return, "Skipping test since /dev/null does not exist");
+		}
+
+		open_fds[i] = fd;
+	}
+
+	pid = sys_clone3(&args, sizeof(args));
+	ASSERT_GE(pid, 0);
+
+	if (pid == 0) {
+		ret = sys_close_range(open_fds[0], UINT_MAX,
+				      CLOSE_RANGE_UNSHARE);
+		if (ret)
+			exit(EXIT_FAILURE);
+
+		for (i = 0; i <= 100; i++)
+			if (fcntl(open_fds[i], F_GETFL) != -1)
+				exit(EXIT_FAILURE);
+
+		exit(EXIT_SUCCESS);
+	}
+
+	EXPECT_EQ(waitpid(pid, &status, 0), pid);
+	EXPECT_EQ(true, WIFEXITED(status));
+	EXPECT_EQ(0, WEXITSTATUS(status));
+}
+
+TEST_HARNESS_MAIN