ARM: 9485/1: mm: acquire mmap write lock around show_pte() for user faultsgrafted

When CONFIG_DEBUG_USER=y, and cmdline "user_debug=31" is set,
a user fault may trigger show_pte() without any lock.
If another thread in the same process concurrently calls munmap(),
the page table pages may be freed while show_pte() is still traversing
them, causing a use-after-free in show_pte().

If CONFIG_ARM_LPAE=y, this may cause a kernel panic if the pages table
of PMD are freed when show_pte() is running.

Acquire mmap_write_lock() around show_pte() for user faults to fix the
contention.

For user faults, additionally restrict that show_pte() is called only
when the addr is a user-space address (addr < TASK_SIZE). This is because
the lock of tsk->mm only protects the virtual memory of user address space,
furthermore, dumping the page tables of a kernel-space address for user
faults is unnecessary and may have security implications.

Keep everything unchanged for kernel faults, because the kernel is
already in the "oops" state, acquiring a lock may risk a deadlock.

Co-developed-by: Qi Xi <xiqi2@huawei.com>

Fixes: 6d021b724481 ("ARM: dump pgd, pmd and pte states on unhandled data abort faults")
Link: https://lore.kernel.org/20260716014022.2823-1-xieyuanbin1@huawei.com
Acked-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
Reviewed-by: Linus Walleij <linusw@kernel.org>
Signed-off-by: Qi Xi <xiqi2@huawei.com>
Signed-off-by: Xie Yuanbin <xieyuanbin1@huawei.com>
Signed-off-by: Russell King <rmk+kernel@armlinux.org.uk>

1 files changed, 92 insertions, 0 deletions

diff --git a/tools/perf/util/cloexec.c b/tools/perf/util/cloexec.c
new file mode 100644
index 000000000..8830604c3
--- /dev/null
+++ b/tools/perf/util/cloexec.c
@@ -0,0 +1,92 @@
+// SPDX-License-Identifier: GPL-2.0
+#include <errno.h>
+#include <sched.h>
+#include "util.h" // for sched_getcpu()
+#include "../perf-sys.h"
+#include "cloexec.h"
+#include "event.h"
+#include "asm/bug.h"
+#include "debug.h"
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <linux/string.h>
+
+static unsigned long flag = PERF_FLAG_FD_CLOEXEC;
+
+static int perf_flag_probe(void)
+{
+	/* use 'safest' configuration as used in evsel__fallback() */
+	struct perf_event_attr attr = {
+		.type = PERF_TYPE_SOFTWARE,
+		.config = PERF_COUNT_SW_CPU_CLOCK,
+		.exclude_kernel = 1,
+	};
+	int fd;
+	int err;
+	int cpu;
+	pid_t pid = -1;
+	char sbuf[STRERR_BUFSIZE];
+
+	cpu = sched_getcpu();
+	if (cpu < 0)
+		cpu = 0;
+
+	/*
+	 * Using -1 for the pid is a workaround to avoid gratuitous jump label
+	 * changes.
+	 */
+	while (1) {
+		/* check cloexec flag */
+		fd = sys_perf_event_open(&attr, pid, cpu, -1,
+					 PERF_FLAG_FD_CLOEXEC);
+		if (fd < 0 && pid == -1 && errno == EACCES) {
+			pid = 0;
+			continue;
+		}
+		break;
+	}
+	err = errno;
+
+	if (fd >= 0) {
+		close(fd);
+		return 1;
+	}
+
+	WARN_ONCE(err != EINVAL && err != EBUSY && err != EACCES,
+		  "perf_event_open(..., PERF_FLAG_FD_CLOEXEC) failed with unexpected error %d (%s)\n",
+		  err, str_error_r(err, sbuf, sizeof(sbuf)));
+
+	/* not supported, confirm error related to PERF_FLAG_FD_CLOEXEC */
+	while (1) {
+		fd = sys_perf_event_open(&attr, pid, cpu, -1, 0);
+		if (fd < 0 && pid == -1 && errno == EACCES) {
+			pid = 0;
+			continue;
+		}
+		break;
+	}
+	err = errno;
+
+	if (fd >= 0)
+		close(fd);
+
+	if (WARN_ONCE(fd < 0 && err != EBUSY && err != EACCES,
+		      "perf_event_open(..., 0) failed unexpectedly with error %d (%s)\n",
+		      err, str_error_r(err, sbuf, sizeof(sbuf))))
+		return -1;
+
+	return 0;
+}
+
+unsigned long perf_event_open_cloexec_flag(void)
+{
+	static bool probed;
+
+	if (!probed) {
+		if (perf_flag_probe() <= 0)
+			flag = 0;
+		probed = true;
+	}
+
+	return flag;
+}