[PATCH v2 0/3] kcov: collect coverage from usb and vhost

[PATCH v2 0/3] kcov: collect coverage from usb and vhost

[Andrey Konovalov]

This patchset extends kcov to allow collecting coverage from the USB
subsystem and vhost workers. See the first patch description for details
about the kcov extension. The other two patches apply this kcov extension
to USB and vhost.

These patches have been used to enable coverage-guided USB fuzzing with
syzkaller for the last few years, see the details here:

https://github.com/google/syzkaller/blob/master/docs/linux/external_fuzzing_usb.md

This patchset has been pushed to the public Linux kernel Gerrit instance:

https://linux-review.googlesource.com/c/linux/kernel/git/torvalds/linux/+/1524

Changes v1 -> v2:
- Changed common_handle type back to u64 (to allow extending it in the
  future).
- Reworked kcov_remote_handle() helpers.
- Fixed vhost annotations when CONFIG_KCOV is not enabled.
- Use kcov_disable() instead of kcov_remote_reset() when
  KCOV_REMOTE_ENABLE fails.

Changes RFC v1 -> v1:
- Remove unnecessary #ifdef's from drivers/vhost/vhost.c.
- Reset t->kcov when area allocation fails in kcov_remote_start().
- Use struct_size to calculate array size in kcov_ioctl().
- Add a limit on area_size in kcov_remote_arg.
- Added kcov_disable() helper.
- Changed encoding of kcov remote handle ids, see the documentation.
- Added a comment reference for kcov_sequence task_struct field.
- Change common_handle type to u32.
- Add checks for handle validity into kcov_ioctl_locked() and
    kcov_remote_start().
- Updated documentation to reflect the changes.

Andrey Konovalov (3):
  kcov: remote coverage support
  usb, kcov: collect coverage from hub_event
  vhost, kcov: collect coverage from vhost_worker

 Documentation/dev-tools/kcov.rst | 128 +++++++++
 drivers/usb/core/hub.c           |   5 +
 drivers/vhost/vhost.c            |   6 +
 drivers/vhost/vhost.h            |   1 +
 include/linux/kcov.h             |  23 ++
 include/linux/sched.h            |   6 +
 include/uapi/linux/kcov.h        |  25 ++
 kernel/kcov.c                    | 480 ++++++++++++++++++++++++++++---
 8 files changed, 639 insertions(+), 35 deletions(-)

-- 
2.24.0.rc0.303.g954a862665-goog

[PATCH v2 1/3] kcov: remote coverage support

[Andrey Konovalov]

This patch adds background thread coverage collection ability to kcov.

With KCOV_ENABLE coverage is collected only for syscalls that are issued
from the current process. With KCOV_REMOTE_ENABLE it's possible to collect
coverage for arbitrary parts of the kernel code, provided that those parts
are annotated with kcov_remote_start()/kcov_remote_stop().

This allows to collect coverage from two types of kernel background
threads: the global ones, that are spawned during kernel boot and are
always running (e.g. USB hub_event()); and the local ones, that are
spawned when a user interacts with some kernel interface (e.g. vhost
workers).

To enable collecting coverage from a global background thread, a unique
global handle must be assigned and passed to the corresponding
kcov_remote_start() call. Then a userspace process can pass a list of such
handles to the KCOV_REMOTE_ENABLE ioctl in the handles array field of the
kcov_remote_arg struct. This will attach the used kcov device to the code
sections, that are referenced by those handles.

Since there might be many local background threads spawned from different
userspace processes, we can't use a single global handle per annotation.
Instead, the userspace process passes a non-zero handle through the
common_handle field of the kcov_remote_arg struct. This common handle gets
saved to the kcov_handle field in the current task_struct and needs to be
passed to the newly spawned threads via custom annotations. Those threads
should in turn be annotated with kcov_remote_start()/kcov_remote_stop().

Internally kcov stores handles as u64 integers. The top byte of a handle
is used to denote the id of a subsystem that this handle belongs to, and
the lower 4 bytes are used to denote a handle id within that subsystem.
A reserved value 0 is used as a subsystem id for common handles as they
don't belong to a particular subsystem. The bytes 4-7 are currently
reserved and must be zero. In the future the number of bytes used for the
subsystem or handle ids might be increased.

When a particular userspace proccess collects coverage by via a common
handle, kcov will collect coverage for each code section that is annotated
to use the common handle obtained as kcov_handle from the current
task_struct. However non common handles allow to collect coverage
selectively from different subsystems.

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
---
 Documentation/dev-tools/kcov.rst | 128 +++++++++
 include/linux/kcov.h             |  23 ++
 include/linux/sched.h            |   6 +
 include/uapi/linux/kcov.h        |  25 ++
 kernel/kcov.c                    | 480 ++++++++++++++++++++++++++++---
 5 files changed, 627 insertions(+), 35 deletions(-)

diff --git a/Documentation/dev-tools/kcov.rst b/Documentation/dev-tools/kcov.rst
index 42b612677799..80536ffe784a 100644
--- a/Documentation/dev-tools/kcov.rst
+++ b/Documentation/dev-tools/kcov.rst
@@ -34,6 +34,7 @@ Profiling data will only become accessible once debugfs has been mounted::
 
 Coverage collection
 -------------------
+
 The following program demonstrates coverage collection from within a test
 program using kcov:
 
@@ -128,6 +129,7 @@ only need to enable coverage (disable happens automatically on thread end).
 
 Comparison operands collection
 ------------------------------
+
 Comparison operands collection is similar to coverage collection:
 
 .. code-block:: c
@@ -202,3 +204,129 @@ Comparison operands collection is similar to coverage collection:
 
 Note that the kcov modes (coverage collection or comparison operands) are
 mutually exclusive.
+
+Remote coverage collection
+--------------------------
+
+With KCOV_ENABLE coverage is collected only for syscalls that are issued from
+the current process. With KCOV_REMOTE_ENABLE it's possible to collect coverage
+for arbitrary parts of the kernel code, provided that those parts are annotated
+with kcov_remote_start()/kcov_remote_stop().
+
+This allows to collect coverage from two types of kernel background threads:
+the global ones, that are spawned during kernel boot and are always running
+(e.g. USB hub_event()); and the local ones, that are spawned when a user
+interacts with some kernel interface (e.g. vhost workers).
+
+To enable collecting coverage from a global background thread, a unique global
+handle must be assigned and passed to the corresponding kcov_remote_start()
+call. Then a userspace process can pass a list of such handles to the
+KCOV_REMOTE_ENABLE ioctl in the handles array field of the kcov_remote_arg
+struct. This will attach the used kcov device to the code sections, that are
+referenced by those handles.
+
+Since there might be many local background threads spawned from different
+userspace processes, we can't use a single global handle per annotation.
+Instead, the userspace process passes a non-zero handle through the
+common_handle field of the kcov_remote_arg struct. This common handle gets
+saved to the kcov_handle field in the current task_struct and needs to be
+passed to the newly spawned threads via custom annotations. Those threads
+should in turn be annotated with kcov_remote_start()/kcov_remote_stop().
+
+Internally kcov stores handles as u64 integers. The top byte of a handle is
+used to denote the id of a subsystem that this handle belongs to, and the lower
+4 bytes are used to denote a handle id within that subsystem. A reserved value
+0 is used as a subsystem id for common handles as they don't belong to a
+particular subsystem. The bytes 4-7 are currently reserved and must be zero.
+In the future the number of bytes used for the subsystem or handle ids might
+be increased.
+
+When a particular userspace proccess collects coverage by via a common handle,
+kcov will collect coverage for each code section that is annotated to use the
+common handle obtained as kcov_handle from the current task_struct. However
+non common handles allow to collect coverage selectively from different
+subsystems.
+
+.. code-block:: c
+
+    struct kcov_remote_arg {
+	unsigned	trace_mode;
+	unsigned	area_size;
+	unsigned	num_handles;
+	uint64_t	common_handle;
+	uint64_t	handles[0];
+    };
+
+    #define KCOV_INIT_TRACE			_IOR('c', 1, unsigned long)
+    #define KCOV_DISABLE			_IO('c', 101)
+    #define KCOV_REMOTE_ENABLE		_IOW('c', 102, struct kcov_remote_arg)
+
+    #define COVER_SIZE	(64 << 10)
+
+    #define KCOV_TRACE_PC	0
+
+    #define KCOV_SUBSYSTEM_COMMON	(0x00ull << 56)
+    #define KCOV_SUBSYSTEM_USB	(0x01ull << 56)
+
+    #define KCOV_SUBSYSTEM_MASK	(0xffull << 56)
+    #define KCOV_HANDLE_ID_MASK	(0xffffffffull)
+
+    static inline __u64 kcov_remote_handle(__u64 subsys, __u64 id)
+    {
+	if (subsys & ~KCOV_SUBSYSTEM_MASK || id & ~KCOV_HANDLE_ID_MASK)
+		return 0;
+	return subsys | id;
+    }
+
+    #define KCOV_COMMON_ID	0x42
+    #define KCOV_USB_BUS_NUM	1
+
+    int main(int argc, char **argv)
+    {
+	int fd;
+	unsigned long *cover, n, i;
+	struct kcov_remote_arg *arg;
+
+	fd = open("/sys/kernel/debug/kcov", O_RDWR);
+	if (fd == -1)
+		perror("open"), exit(1);
+	if (ioctl(fd, KCOV_INIT_TRACE, COVER_SIZE))
+		perror("ioctl"), exit(1);
+	cover = (unsigned long*)mmap(NULL, COVER_SIZE * sizeof(unsigned long),
+				     PROT_READ | PROT_WRITE, MAP_SHARED, fd, 0);
+	if ((void*)cover == MAP_FAILED)
+		perror("mmap"), exit(1);
+
+	/* Enable coverage collection via common handle and from USB bus #1. */
+	arg = calloc(1, sizeof(*arg) + sizeof(uint64_t));
+	if (!arg)
+		perror("calloc"), exit(1);
+	arg->trace_mode = KCOV_TRACE_PC;
+	arg->area_size = COVER_SIZE;
+	arg->num_handles = 1;
+	arg->common_handle = kcov_remote_handle(KCOV_SUBSYSTEM_COMMON,
+							KCOV_COMMON_ID);
+	arg->handles[0] = kcov_remote_handle(KCOV_SUBSYSTEM_USB,
+						KCOV_USB_BUS_NUM);
+	if (ioctl(fd, KCOV_REMOTE_ENABLE, arg))
+		perror("ioctl"), free(arg), exit(1);
+	free(arg);
+
+	/*
+	 * Here the user needs to trigger execution of a kernel code section
+	 * that is either annotated with the common handle, or to trigger some
+	 * activity on USB bus #1.
+	 */
+	sleep(2);
+
+	n = __atomic_load_n(&cover[0], __ATOMIC_RELAXED);
+	for (i = 0; i < n; i++)
+		printf("0x%lx\n", cover[i + 1]);
+	if (ioctl(fd, KCOV_DISABLE, 0))
+		perror("ioctl"), exit(1);
+	if (munmap(cover, COVER_SIZE * sizeof(unsigned long)))
+		perror("munmap"), exit(1);
+	if (close(fd))
+		perror("close"), exit(1);
+	return 0;
+    }
diff --git a/include/linux/kcov.h b/include/linux/kcov.h
index b76a1807028d..a10e84707d82 100644
--- a/include/linux/kcov.h
+++ b/include/linux/kcov.h
@@ -37,12 +37,35 @@ do {						\
 	(t)->kcov_mode &= ~KCOV_IN_CTXSW;	\
 } while (0)
 
+/* See Documentation/dev-tools/kcov.rst for usage details. */
+void kcov_remote_start(u64 handle);
+void kcov_remote_stop(void);
+u64 kcov_common_handle(void);
+
+static inline void kcov_remote_start_common(u64 id)
+{
+	kcov_remote_start(kcov_remote_handle(KCOV_SUBSYSTEM_COMMON, id));
+}
+
+static inline void kcov_remote_start_usb(u64 id)
+{
+	kcov_remote_start(kcov_remote_handle(KCOV_SUBSYSTEM_USB, id));
+}
+
 #else
 
 static inline void kcov_task_init(struct task_struct *t) {}
 static inline void kcov_task_exit(struct task_struct *t) {}
 static inline void kcov_prepare_switch(struct task_struct *t) {}
 static inline void kcov_finish_switch(struct task_struct *t) {}
+static inline void kcov_remote_start(u64 handle) {}
+static inline void kcov_remote_stop(void) {}
+static inline u64 kcov_common_handle(void)
+{
+	return 0;
+}
+static inline void kcov_remote_start_common(u64 id) {}
+static inline void kcov_remote_start_usb(u64 id) {}
 
 #endif /* CONFIG_KCOV */
 #endif /* _LINUX_KCOV_H */
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 67a1d86981a9..b0f9534b971e 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -1214,6 +1214,12 @@ struct task_struct {
 
 	/* KCOV descriptor wired with this task or NULL: */
 	struct kcov			*kcov;
+
+	/* KCOV common handle for remote coverage collection: */
+	u64				kcov_handle;
+
+	/* KCOV sequence number, see kernel/kcov.c for details: */
+	int				kcov_sequence;
 #endif
 
 #ifdef CONFIG_MEMCG
diff --git a/include/uapi/linux/kcov.h b/include/uapi/linux/kcov.h
index 9529867717a8..1236847b2356 100644
--- a/include/uapi/linux/kcov.h
+++ b/include/uapi/linux/kcov.h
@@ -4,9 +4,21 @@
 
 #include <linux/types.h>
 
+/* See Documentation/dev-tools/kcov.rst for usage details. */
+struct kcov_remote_arg {
+	unsigned int	trace_mode;	/* KCOV_TRACE_PC or KCOV_TRACE_CMP */
+	unsigned int	area_size;	/* Length of coverage buffer in words */
+	unsigned int	num_handles;	/* Size of handles array */
+	__u64		common_handle;
+	__u64		handles[0];
+};
+
+#define KCOV_REMOTE_MAX_HANDLES		0x10000
+
 #define KCOV_INIT_TRACE			_IOR('c', 1, unsigned long)
 #define KCOV_ENABLE			_IO('c', 100)
 #define KCOV_DISABLE			_IO('c', 101)
+#define KCOV_REMOTE_ENABLE		_IOW('c', 102, struct kcov_remote_arg)
 
 enum {
 	/*
@@ -32,4 +44,17 @@ enum {
 #define KCOV_CMP_SIZE(n)        ((n) << 1)
 #define KCOV_CMP_MASK           KCOV_CMP_SIZE(3)
 
+#define KCOV_SUBSYSTEM_COMMON	(0x00ull << 56)
+#define KCOV_SUBSYSTEM_USB	(0x01ull << 56)
+
+#define KCOV_SUBSYSTEM_MASK	(0xffull << 56)
+#define KCOV_HANDLE_ID_MASK	(0xffffffffull)
+
+static inline __u64 kcov_remote_handle(__u64 subsys, __u64 id)
+{
+	if (subsys & ~KCOV_SUBSYSTEM_MASK || id & ~KCOV_HANDLE_ID_MASK)
+		return 0;
+	return subsys | id;
+}
+
 #endif /* _LINUX_KCOV_IOCTLS_H */
diff --git a/kernel/kcov.c b/kernel/kcov.c
index 2ee38727844a..a79d2351212b 100644
--- a/kernel/kcov.c
+++ b/kernel/kcov.c
@@ -9,6 +9,7 @@
 #include <linux/types.h>
 #include <linux/file.h>
 #include <linux/fs.h>
+#include <linux/hashtable.h>
 #include <linux/init.h>
 #include <linux/mm.h>
 #include <linux/preempt.h>
@@ -23,6 +24,8 @@
 #include <linux/refcount.h>
 #include <asm/setup.h>
 
+#define kcov_debug(fmt, ...) pr_debug("%s: " fmt, __func__, ##__VA_ARGS__)
+
 /* Number of 64-bit words written per one comparison: */
 #define KCOV_WORDS_PER_CMP 4
 
@@ -44,19 +47,97 @@ struct kcov {
 	 * Reference counter. We keep one for:
 	 *  - opened file descriptor
 	 *  - task with enabled coverage (we can't unwire it from another task)
+	 *  - each code section for remote coverage collection
 	 */
 	refcount_t		refcount;
 	/* The lock protects mode, size, area and t. */
 	spinlock_t		lock;
 	enum kcov_mode		mode;
-	/* Size of arena (in long's for KCOV_MODE_TRACE). */
-	unsigned		size;
+	/* Size of arena (in long's). */
+	unsigned int		size;
 	/* Coverage buffer shared with user space. */
 	void			*area;
 	/* Task for which we collect coverage, or NULL. */
 	struct task_struct	*t;
+	/* Collecting coverage from remote threads. */
+	bool			remote;
+	/* Size of remote arena (in long's). */
+	unsigned int		remote_size;
+	/*
+	 * Sequence is incremented each time kcov is reenabled, used by
+	 * kcov_remote_stop(), see the comment there.
+	 */
+	int			sequence;
 };
 
+struct kcov_remote_area {
+	struct list_head	list;
+	unsigned int		size;
+};
+
+struct kcov_remote {
+	u64			handle;
+	struct kcov		*kcov;
+	struct hlist_node	hnode;
+};
+
+static DEFINE_SPINLOCK(kcov_remote_lock);
+static DEFINE_HASHTABLE(kcov_remote_map, 4);
+static struct list_head kcov_remote_areas = LIST_HEAD_INIT(kcov_remote_areas);
+
+static struct kcov_remote *kcov_remote_find(u64 handle)
+{
+	struct kcov_remote *remote;
+
+	hash_for_each_possible(kcov_remote_map, remote, hnode, handle) {
+		if (remote->handle == handle)
+			return remote;
+	}
+	return NULL;
+}
+
+static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
+{
+	struct kcov_remote *remote;
+
+	if (kcov_remote_find(handle))
+		return ERR_PTR(-EEXIST);
+	remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
+	if (!remote)
+		return ERR_PTR(-ENOMEM);
+	remote->handle = handle;
+	remote->kcov = kcov;
+	hash_add(kcov_remote_map, &remote->hnode, handle);
+	return remote;
+}
+
+static struct kcov_remote_area *kcov_remote_area_get(unsigned int size)
+{
+	struct kcov_remote_area *area;
+	struct list_head *pos;
+
+	kcov_debug("size = %u\n", size);
+	list_for_each(pos, &kcov_remote_areas) {
+		area = list_entry(pos, struct kcov_remote_area, list);
+		if (area->size == size) {
+			list_del(&area->list);
+			kcov_debug("rv = %px\n", area);
+			return area;
+		}
+	}
+	kcov_debug("rv = NULL\n");
+	return NULL;
+}
+
+static void kcov_remote_area_put(struct kcov_remote_area *area,
+					unsigned int size)
+{
+	kcov_debug("area = %px, size = %u\n", area, size);
+	INIT_LIST_HEAD(&area->list);
+	area->size = size;
+	list_add(&area->list, &kcov_remote_areas);
+}
+
 static notrace bool check_kcov_mode(enum kcov_mode needed_mode, struct task_struct *t)
 {
 	unsigned int mode;
@@ -73,7 +154,7 @@ static notrace bool check_kcov_mode(enum kcov_mode needed_mode, struct task_stru
 	 * in_interrupt() returns false (e.g. preempt_schedule_irq()).
 	 * READ_ONCE()/barrier() effectively provides load-acquire wrt
 	 * interrupts, there are paired barrier()/WRITE_ONCE() in
-	 * kcov_ioctl_locked().
+	 * kcov_start().
 	 */
 	barrier();
 	return mode == needed_mode;
@@ -227,6 +308,78 @@ void notrace __sanitizer_cov_trace_switch(u64 val, u64 *cases)
 EXPORT_SYMBOL(__sanitizer_cov_trace_switch);
 #endif /* ifdef CONFIG_KCOV_ENABLE_COMPARISONS */
 
+static void kcov_start(struct task_struct *t, unsigned int size,
+			void *area, enum kcov_mode mode, int sequence)
+{
+	kcov_debug("t = %px, size = %u, area = %px\n", t, size, area);
+	/* Cache in task struct for performance. */
+	t->kcov_size = size;
+	t->kcov_area = area;
+	/* See comment in check_kcov_mode(). */
+	barrier();
+	WRITE_ONCE(t->kcov_mode, mode);
+	t->kcov_sequence = sequence;
+}
+
+static void kcov_stop(struct task_struct *t)
+{
+	WRITE_ONCE(t->kcov_mode, KCOV_MODE_DISABLED);
+	barrier();
+	t->kcov_size = 0;
+	t->kcov_area = NULL;
+}
+
+static void kcov_task_reset(struct task_struct *t)
+{
+	kcov_stop(t);
+	t->kcov = NULL;
+	t->kcov_sequence = 0;
+	t->kcov_handle = 0;
+}
+
+void kcov_task_init(struct task_struct *t)
+{
+	kcov_task_reset(t);
+	t->kcov_handle = current->kcov_handle;
+}
+
+static void kcov_reset(struct kcov *kcov)
+{
+	kcov->t = NULL;
+	kcov->mode = KCOV_MODE_INIT;
+	kcov->remote = false;
+	kcov->remote_size = 0;
+	kcov->sequence++;
+}
+
+static void kcov_remote_reset(struct kcov *kcov)
+{
+	int bkt;
+	struct kcov_remote *remote;
+	struct hlist_node *tmp;
+
+	spin_lock(&kcov_remote_lock);
+	hash_for_each_safe(kcov_remote_map, bkt, tmp, remote, hnode) {
+		if (remote->kcov != kcov)
+			continue;
+		kcov_debug("removing handle %llx\n", remote->handle);
+		hash_del(&remote->hnode);
+		kfree(remote);
+	}
+	/* Do reset before unlock to prevent races with kcov_remote_start(). */
+	kcov_reset(kcov);
+	spin_unlock(&kcov_remote_lock);
+}
+
+static void kcov_disable(struct task_struct *t, struct kcov *kcov)
+{
+	kcov_task_reset(t);
+	if (kcov->remote)
+		kcov_remote_reset(kcov);
+	else
+		kcov_reset(kcov);
+}
+
 static void kcov_get(struct kcov *kcov)
 {
 	refcount_inc(&kcov->refcount);
@@ -235,20 +388,12 @@ static void kcov_get(struct kcov *kcov)
 static void kcov_put(struct kcov *kcov)
 {
 	if (refcount_dec_and_test(&kcov->refcount)) {
+		kcov_remote_reset(kcov);
 		vfree(kcov->area);
 		kfree(kcov);
 	}
 }
 
-void kcov_task_init(struct task_struct *t)
-{
-	WRITE_ONCE(t->kcov_mode, KCOV_MODE_DISABLED);
-	barrier();
-	t->kcov_size = 0;
-	t->kcov_area = NULL;
-	t->kcov = NULL;
-}
-
 void kcov_task_exit(struct task_struct *t)
 {
 	struct kcov *kcov;
@@ -256,15 +401,23 @@ void kcov_task_exit(struct task_struct *t)
 	kcov = t->kcov;
 	if (kcov == NULL)
 		return;
+
 	spin_lock(&kcov->lock);
+	kcov_debug("t = %px, kcov->t = %px\n", t, kcov->t);
+	/*
+	 * If !kcov->remote, this checks that t->kcov->t == t.
+	 * If kcov->remote == true then the exiting task is either:
+	 * 1. a remote task between kcov_remote_start() and kcov_remote_stop(),
+	 *    in this case t != kcov->t and we'll print a warning; or
+	 * 2. the task that created kcov exiting without calling KCOV_DISABLE,
+	 *    in this case t == kcov->t and no warning is printed.
+	 */
 	if (WARN_ON(kcov->t != t)) {
 		spin_unlock(&kcov->lock);
 		return;
 	}
 	/* Just to not leave dangling references behind. */
-	kcov_task_init(t);
-	kcov->t = NULL;
-	kcov->mode = KCOV_MODE_INIT;
+	kcov_disable(t, kcov);
 	spin_unlock(&kcov->lock);
 	kcov_put(kcov);
 }
@@ -313,6 +466,7 @@ static int kcov_open(struct inode *inode, struct file *filep)
 	if (!kcov)
 		return -ENOMEM;
 	kcov->mode = KCOV_MODE_DISABLED;
+	kcov->sequence = 1;
 	refcount_set(&kcov->refcount, 1);
 	spin_lock_init(&kcov->lock);
 	filep->private_data = kcov;
@@ -325,6 +479,20 @@ static int kcov_close(struct inode *inode, struct file *filep)
 	return 0;
 }
 
+static int kcov_get_mode(unsigned long arg)
+{
+	if (arg == KCOV_TRACE_PC)
+		return KCOV_MODE_TRACE_PC;
+	else if (arg == KCOV_TRACE_CMP)
+#ifdef CONFIG_KCOV_ENABLE_COMPARISONS
+		return KCOV_MODE_TRACE_CMP;
+#else
+		return -ENOTSUPP;
+#endif
+	else
+		return -EINVAL;
+}
+
 /*
  * Fault in a lazily-faulted vmalloc area before it can be used by
  * __santizer_cov_trace_pc(), to avoid recursion issues if any code on the
@@ -340,14 +508,35 @@ static void kcov_fault_in_area(struct kcov *kcov)
 		READ_ONCE(area[offset]);
 }
 
+static inline bool kcov_check_handle(u64 handle, bool common_valid,
+				bool uncommon_valid, bool zero_valid)
+{
+	if (handle & ~(KCOV_SUBSYSTEM_MASK | KCOV_HANDLE_ID_MASK))
+		return false;
+	switch (handle & KCOV_SUBSYSTEM_MASK) {
+	case KCOV_SUBSYSTEM_COMMON:
+		return (handle & KCOV_HANDLE_ID_MASK) ?
+			common_valid : zero_valid;
+	case KCOV_SUBSYSTEM_USB:
+		return uncommon_valid;
+	default:
+		return false;
+	}
+	return false;
+}
+
 static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
 			     unsigned long arg)
 {
 	struct task_struct *t;
 	unsigned long size, unused;
+	int mode, i;
+	struct kcov_remote_arg *remote_arg;
+	struct kcov_remote *remote;
 
 	switch (cmd) {
 	case KCOV_INIT_TRACE:
+		kcov_debug("KCOV_INIT_TRACE\n");
 		/*
 		 * Enable kcov in trace mode and setup buffer size.
 		 * Must happen before anything else.
@@ -366,6 +555,7 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
 		kcov->mode = KCOV_MODE_INIT;
 		return 0;
 	case KCOV_ENABLE:
+		kcov_debug("KCOV_ENABLE\n");
 		/*
 		 * Enable coverage for the current task.
 		 * At this point user must have been enabled trace mode,
@@ -378,29 +568,20 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
 		t = current;
 		if (kcov->t != NULL || t->kcov != NULL)
 			return -EBUSY;
-		if (arg == KCOV_TRACE_PC)
-			kcov->mode = KCOV_MODE_TRACE_PC;
-		else if (arg == KCOV_TRACE_CMP)
-#ifdef CONFIG_KCOV_ENABLE_COMPARISONS
-			kcov->mode = KCOV_MODE_TRACE_CMP;
-#else
-		return -ENOTSUPP;
-#endif
-		else
-			return -EINVAL;
+		mode = kcov_get_mode(arg);
+		if (mode < 0)
+			return mode;
 		kcov_fault_in_area(kcov);
-		/* Cache in task struct for performance. */
-		t->kcov_size = kcov->size;
-		t->kcov_area = kcov->area;
-		/* See comment in check_kcov_mode(). */
-		barrier();
-		WRITE_ONCE(t->kcov_mode, kcov->mode);
+		kcov->mode = mode;
+		kcov_start(t, kcov->size, kcov->area, kcov->mode,
+				kcov->sequence);
 		t->kcov = kcov;
 		kcov->t = t;
-		/* This is put either in kcov_task_exit() or in KCOV_DISABLE. */
+		/* Put either in kcov_task_exit() or in KCOV_DISABLE. */
 		kcov_get(kcov);
 		return 0;
 	case KCOV_DISABLE:
+		kcov_debug("KCOV_DISABLE\n");
 		/* Disable coverage for the current task. */
 		unused = arg;
 		if (unused != 0 || current->kcov != kcov)
@@ -408,11 +589,65 @@ static int kcov_ioctl_locked(struct kcov *kcov, unsigned int cmd,
 		t = current;
 		if (WARN_ON(kcov->t != t))
 			return -EINVAL;
-		kcov_task_init(t);
-		kcov->t = NULL;
-		kcov->mode = KCOV_MODE_INIT;
+		kcov_disable(t, kcov);
 		kcov_put(kcov);
 		return 0;
+	case KCOV_REMOTE_ENABLE:
+		kcov_debug("KCOV_REMOTE_ENABLE\n");
+		if (kcov->mode != KCOV_MODE_INIT || !kcov->area)
+			return -EINVAL;
+		t = current;
+		if (kcov->t != NULL || t->kcov != NULL)
+			return -EBUSY;
+		remote_arg = (struct kcov_remote_arg *)arg;
+		mode = kcov_get_mode(remote_arg->trace_mode);
+		if (mode < 0)
+			return mode;
+		if (remote_arg->area_size > LONG_MAX / sizeof(unsigned long))
+			return -EINVAL;
+		kcov->mode = mode;
+		t->kcov = kcov;
+		kcov->t = t;
+		kcov->remote = true;
+		kcov->remote_size = remote_arg->area_size;
+		spin_lock(&kcov_remote_lock);
+		for (i = 0; i < remote_arg->num_handles; i++) {
+			kcov_debug("handle %llx\n", remote_arg->handles[i]);
+			if (!kcov_check_handle(remote_arg->handles[i],
+						false, true, false)) {
+				spin_unlock(&kcov_remote_lock);
+				kcov_disable(t, kcov);
+				return -EINVAL;
+			}
+			remote = kcov_remote_add(kcov, remote_arg->handles[i]);
+			if (IS_ERR(remote)) {
+				spin_unlock(&kcov_remote_lock);
+				kcov_disable(t, kcov);
+				return PTR_ERR(remote);
+			}
+		}
+		if (remote_arg->common_handle) {
+			kcov_debug("common handle %llx\n",
+					remote_arg->common_handle);
+			if (!kcov_check_handle(remote_arg->common_handle,
+						true, false, false)) {
+				spin_unlock(&kcov_remote_lock);
+				kcov_disable(t, kcov);
+				return -EINVAL;
+			}
+			remote = kcov_remote_add(kcov,
+					remote_arg->common_handle);
+			if (IS_ERR(remote)) {
+				spin_unlock(&kcov_remote_lock);
+				kcov_disable(t, kcov);
+				return PTR_ERR(remote);
+			}
+			t->kcov_handle = remote_arg->common_handle;
+		}
+		spin_unlock(&kcov_remote_lock);
+		/* Put either in kcov_task_exit() or in KCOV_DISABLE. */
+		kcov_get(kcov);
+		return 0;
 	default:
 		return -ENOTTY;
 	}
@@ -422,11 +657,35 @@ static long kcov_ioctl(struct file *filep, unsigned int cmd, unsigned long arg)
 {
 	struct kcov *kcov;
 	int res;
+	struct kcov_remote_arg *remote_arg = NULL;
+	unsigned int remote_num_handles;
+	unsigned long remote_arg_size;
+
+	if (cmd == KCOV_REMOTE_ENABLE) {
+		if (get_user(remote_num_handles, (unsigned __user *)(arg +
+				offsetof(struct kcov_remote_arg, num_handles))))
+			return -EFAULT;
+		if (remote_num_handles > KCOV_REMOTE_MAX_HANDLES)
+			return -EINVAL;
+		remote_arg_size = struct_size(remote_arg, handles,
+					remote_num_handles);
+		remote_arg = memdup_user((void __user *)arg, remote_arg_size);
+		if (IS_ERR(remote_arg))
+			return PTR_ERR(remote_arg);
+		if (remote_arg->num_handles != remote_num_handles) {
+			kfree(remote_arg);
+			return -EINVAL;
+		}
+		arg = (unsigned long)remote_arg;
+	}
 
 	kcov = filep->private_data;
 	spin_lock(&kcov->lock);
 	res = kcov_ioctl_locked(kcov, cmd, arg);
 	spin_unlock(&kcov->lock);
+
+	kfree(remote_arg);
+
 	return res;
 }
 
@@ -438,6 +697,157 @@ static const struct file_operations kcov_fops = {
 	.release        = kcov_close,
 };
 
+u64 kcov_common_handle(void)
+{
+	return current->kcov_handle;
+}
+
+void kcov_remote_start(u64 handle)
+{
+	struct kcov_remote *remote;
+	void *area;
+	struct task_struct *t;
+	unsigned int size;
+	enum kcov_mode mode;
+	int sequence;
+
+	if (WARN_ON(!kcov_check_handle(handle, true, true, true)))
+		return;
+	if (WARN_ON(!in_task()))
+		return;
+	t = current;
+	/*
+	 * Check that kcov_remote_start is not called twice
+	 * nor called by user tasks (with enabled kcov).
+	 */
+	if (WARN_ON(t->kcov))
+		return;
+
+	kcov_debug("handle = %llx\n", handle);
+
+	spin_lock(&kcov_remote_lock);
+	remote = kcov_remote_find(handle);
+	if (!remote) {
+		kcov_debug("no remote found");
+		spin_unlock(&kcov_remote_lock);
+		return;
+	}
+	/* Put in kcov_remote_stop(). */
+	kcov_get(remote->kcov);
+	t->kcov = remote->kcov;
+	/*
+	 * Read kcov fields before unlock to prevent races with
+	 * KCOV_DISABLE / kcov_remote_reset().
+	 */
+	size = remote->kcov->remote_size;
+	mode = remote->kcov->mode;
+	sequence = remote->kcov->sequence;
+	area = kcov_remote_area_get(size);
+	spin_unlock(&kcov_remote_lock);
+
+	if (!area) {
+		area = vmalloc(size * sizeof(unsigned long));
+		if (!area) {
+			t->kcov = NULL;
+			kcov_put(remote->kcov);
+			return;
+		}
+	}
+	/* Reset coverage size. */
+	*(u64 *)area = 0;
+
+	kcov_debug("area = %px, size = %u", area, size);
+
+	kcov_start(t, size, area, mode, sequence);
+
+}
+
+static void kcov_move_area(enum kcov_mode mode, void *dst_area,
+				unsigned int dst_area_size, void *src_area)
+{
+	u64 word_size = sizeof(unsigned long);
+	u64 count_size, entry_size;
+	u64 dst_len, src_len;
+	void *dst_entries, *src_entries;
+	u64 dst_occupied, dst_free, bytes_to_move, entries_moved;
+
+	kcov_debug("%px %u <= %px %lu\n",
+		dst_area, dst_area_size, src_area, *(unsigned long *)src_area);
+
+	switch (mode) {
+	case KCOV_MODE_TRACE_PC:
+		dst_len = READ_ONCE(*(unsigned long *)dst_area);
+		src_len = *(unsigned long *)src_area;
+		count_size = sizeof(unsigned long);
+		entry_size = sizeof(unsigned long);
+		break;
+	case KCOV_MODE_TRACE_CMP:
+		dst_len = READ_ONCE(*(u64 *)dst_area);
+		src_len = *(u64 *)src_area;
+		count_size = sizeof(u64);
+		entry_size = sizeof(u64) * KCOV_WORDS_PER_CMP;
+		break;
+	default:
+		WARN_ON(1);
+		return;
+	}
+
+	if (dst_len > (dst_area_size * word_size - count_size) / entry_size)
+		return;
+	dst_occupied = count_size + dst_len * entry_size;
+	dst_free = dst_area_size * word_size - dst_occupied;
+	bytes_to_move = min(dst_free, src_len * entry_size);
+	dst_entries = dst_area + dst_occupied;
+	src_entries = src_area + count_size;
+	memcpy(dst_entries, src_entries, bytes_to_move);
+	entries_moved = bytes_to_move / entry_size;
+
+	switch (mode) {
+	case KCOV_MODE_TRACE_PC:
+		WRITE_ONCE(*(unsigned long *)dst_area, dst_len + entries_moved);
+		break;
+	case KCOV_MODE_TRACE_CMP:
+		WRITE_ONCE(*(u64 *)dst_area, dst_len + entries_moved);
+		break;
+	default:
+		break;
+	}
+}
+
+void kcov_remote_stop(void)
+{
+	struct task_struct *t = current;
+	struct kcov *kcov = t->kcov;
+	void *area = t->kcov_area;
+	unsigned int size = t->kcov_size;
+	int sequence = t->kcov_sequence;
+
+	if (!kcov) {
+		kcov_debug("no kcov found\n");
+		return;
+	}
+
+	kcov_stop(t);
+	t->kcov = NULL;
+
+	spin_lock(&kcov->lock);
+	/*
+	 * KCOV_DISABLE could have been called between kcov_remote_start()
+	 * and kcov_remote_stop(), hence the check.
+	 */
+	kcov_debug("move if: %d == %d && %d\n",
+		sequence, kcov->sequence, (int)kcov->remote);
+	if (sequence == kcov->sequence && kcov->remote)
+		kcov_move_area(kcov->mode, kcov->area, kcov->size, area);
+	spin_unlock(&kcov->lock);
+
+	spin_lock(&kcov_remote_lock);
+	kcov_remote_area_put(area, size);
+	spin_unlock(&kcov_remote_lock);
+
+	kcov_put(kcov);
+}
+
 static int __init kcov_init(void)
 {
 	/*
-- 
2.24.0.rc0.303.g954a862665-goog

[PATCH v2 2/3] usb, kcov: collect coverage from hub_event

[Andrey Konovalov]

This patch adds kcov_remote_start()/kcov_remote_stop() annotations to the
hub_event() function, which is responsible for processing events on USB
buses, in particular events that happen during USB device enumeration.
Since hub_event() is run in a global background kernel thread (see
Documentation/dev-tools/kcov.rst for details), each USB bus gets a unique
global handle id from the USB subsystem kcov handle id range. As the
result kcov can now be used to collect coverage from events that happen on
a particular USB bus.

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
---
 drivers/usb/core/hub.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/drivers/usb/core/hub.c b/drivers/usb/core/hub.c
index 236313f41f4a..823dd675f6db 100644
--- a/drivers/usb/core/hub.c
+++ b/drivers/usb/core/hub.c
@@ -29,6 +29,7 @@
 #include <linux/random.h>
 #include <linux/pm_qos.h>
 #include <linux/kobject.h>
+#include <linux/kcov.h>
 
 #include <linux/uaccess.h>
 #include <asm/byteorder.h>
@@ -5374,6 +5375,8 @@ static void hub_event(struct work_struct *work)
 	hub_dev = hub->intfdev;
 	intf = to_usb_interface(hub_dev);
 
+	kcov_remote_start_usb((u64)hdev->bus->busnum);
+
 	dev_dbg(hub_dev, "state %d ports %d chg %04x evt %04x\n",
 			hdev->state, hdev->maxchild,
 			/* NOTE: expects max 15 ports... */
@@ -5480,6 +5483,8 @@ static void hub_event(struct work_struct *work)
 	/* Balance the stuff in kick_hub_wq() and allow autosuspend */
 	usb_autopm_put_interface(intf);
 	kref_put(&hub->kref, hub_release);
+
+	kcov_remote_stop();
 }
 
 static const struct usb_device_id hub_id_table[] = {
-- 
2.24.0.rc0.303.g954a862665-goog

[PATCH v2 3/3] vhost, kcov: collect coverage from vhost_worker

[Andrey Konovalov]

This patch adds kcov_remote_start()/kcov_remote_stop() annotations to the
vhost_worker() function, which is responsible for processing vhost works.
Since vhost_worker() threads are spawned per vhost device instance
the common kcov handle is used for kcov_remote_start()/stop() annotations
(see Documentation/dev-tools/kcov.rst for details). As the result kcov can
now be used to collect coverage from vhost worker threads.

Signed-off-by: Andrey Konovalov <andreyknvl@google.com>
---
 drivers/vhost/vhost.c | 6 ++++++
 drivers/vhost/vhost.h | 1 +
 2 files changed, 7 insertions(+)

diff --git a/drivers/vhost/vhost.c b/drivers/vhost/vhost.c
index 36ca2cf419bf..f44340b41494 100644
--- a/drivers/vhost/vhost.c
+++ b/drivers/vhost/vhost.c
@@ -30,6 +30,7 @@
 #include <linux/sched/signal.h>
 #include <linux/interval_tree_generic.h>
 #include <linux/nospec.h>
+#include <linux/kcov.h>
 
 #include "vhost.h"
 
@@ -357,7 +358,9 @@ static int vhost_worker(void *data)
 		llist_for_each_entry_safe(work, work_next, node, node) {
 			clear_bit(VHOST_WORK_QUEUED, &work->flags);
 			__set_current_state(TASK_RUNNING);
+			kcov_remote_start_common(dev->kcov_handle);
 			work->fn(work);
+			kcov_remote_stop();
 			if (need_resched())
 				schedule();
 		}
@@ -546,6 +549,7 @@ long vhost_dev_set_owner(struct vhost_dev *dev)
 
 	/* No owner, become one */
 	dev->mm = get_task_mm(current);
+	dev->kcov_handle = kcov_common_handle();
 	worker = kthread_create(vhost_worker, dev, "vhost-%d", current->pid);
 	if (IS_ERR(worker)) {
 		err = PTR_ERR(worker);
@@ -571,6 +575,7 @@ long vhost_dev_set_owner(struct vhost_dev *dev)
 	if (dev->mm)
 		mmput(dev->mm);
 	dev->mm = NULL;
+	dev->kcov_handle = 0;
 err_mm:
 	return err;
 }
@@ -682,6 +687,7 @@ void vhost_dev_cleanup(struct vhost_dev *dev)
 	if (dev->worker) {
 		kthread_stop(dev->worker);
 		dev->worker = NULL;
+		dev->kcov_handle = 0;
 	}
 	if (dev->mm)
 		mmput(dev->mm);
diff --git a/drivers/vhost/vhost.h b/drivers/vhost/vhost.h
index e9ed2722b633..a123fd70847e 100644
--- a/drivers/vhost/vhost.h
+++ b/drivers/vhost/vhost.h
@@ -173,6 +173,7 @@ struct vhost_dev {
 	int iov_limit;
 	int weight;
 	int byte_weight;
+	u64 kcov_handle;
 };
 
 bool vhost_exceeds_weight(struct vhost_virtqueue *vq, int pkts, int total_len);
-- 
2.24.0.rc0.303.g954a862665-goog

Re: [PATCH v2 0/3] kcov: collect coverage from usb and vhost

[Andrew Morton]

On Wed, 23 Oct 2019 17:24:28 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:

> This patchset extends kcov to allow collecting coverage from the USB
> subsystem and vhost workers. See the first patch description for details
> about the kcov extension. The other two patches apply this kcov extension
> to USB and vhost.
> 
> These patches have been used to enable coverage-guided USB fuzzing with
> syzkaller for the last few years

I find it surprising that this material is so focused on USB.  Is
there something unique about USB that gave rise to this situation, or
is it expected that the new kcov feature will be used elsewhere in the
kernel?

If the latter, which are the expected subsystems?

Re: [PATCH v2 1/3] kcov: remote coverage support

[Andrew Morton]

On Wed, 23 Oct 2019 17:24:29 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:

> This patch adds background thread coverage collection ability to kcov.
> 
> With KCOV_ENABLE coverage is collected only for syscalls that are issued
> from the current process. With KCOV_REMOTE_ENABLE it's possible to collect
> coverage for arbitrary parts of the kernel code, provided that those parts
> are annotated with kcov_remote_start()/kcov_remote_stop().
> 
> This allows to collect coverage from two types of kernel background
> threads: the global ones, that are spawned during kernel boot and are
> always running (e.g. USB hub_event()); and the local ones, that are
> spawned when a user interacts with some kernel interface (e.g. vhost
> workers).
> 
> To enable collecting coverage from a global background thread, a unique
> global handle must be assigned and passed to the corresponding
> kcov_remote_start() call. Then a userspace process can pass a list of such
> handles to the KCOV_REMOTE_ENABLE ioctl in the handles array field of the
> kcov_remote_arg struct. This will attach the used kcov device to the code
> sections, that are referenced by those handles.
> 
> Since there might be many local background threads spawned from different
> userspace processes, we can't use a single global handle per annotation.
> Instead, the userspace process passes a non-zero handle through the
> common_handle field of the kcov_remote_arg struct. This common handle gets
> saved to the kcov_handle field in the current task_struct and needs to be
> passed to the newly spawned threads via custom annotations. Those threads
> should in turn be annotated with kcov_remote_start()/kcov_remote_stop().
> 
> Internally kcov stores handles as u64 integers. The top byte of a handle
> is used to denote the id of a subsystem that this handle belongs to, and
> the lower 4 bytes are used to denote a handle id within that subsystem.
> A reserved value 0 is used as a subsystem id for common handles as they
> don't belong to a particular subsystem. The bytes 4-7 are currently
> reserved and must be zero. In the future the number of bytes used for the
> subsystem or handle ids might be increased.
> 
> When a particular userspace proccess collects coverage by via a common
> handle, kcov will collect coverage for each code section that is annotated
> to use the common handle obtained as kcov_handle from the current
> task_struct. However non common handles allow to collect coverage
> selectively from different subsystems.
> 
> ...
>
> +static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
> +{
> +	struct kcov_remote *remote;
> +
> +	if (kcov_remote_find(handle))
> +		return ERR_PTR(-EEXIST);
> +	remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
> +	if (!remote)
> +		return ERR_PTR(-ENOMEM);
> +	remote->handle = handle;
> +	remote->kcov = kcov;
> +	hash_add(kcov_remote_map, &remote->hnode, handle);
> +	return remote;
> +}
> +
>
> ...
>
> +		spin_lock(&kcov_remote_lock);
> +		for (i = 0; i < remote_arg->num_handles; i++) {
> +			kcov_debug("handle %llx\n", remote_arg->handles[i]);
> +			if (!kcov_check_handle(remote_arg->handles[i],
> +						false, true, false)) {
> +				spin_unlock(&kcov_remote_lock);
> +				kcov_disable(t, kcov);
> +				return -EINVAL;
> +			}
> +			remote = kcov_remote_add(kcov, remote_arg->handles[i]);
> +			if (IS_ERR(remote)) {
> +				spin_unlock(&kcov_remote_lock);
> +				kcov_disable(t, kcov);
> +				return PTR_ERR(remote);
> +			}
> +		}

It's worrisome that this code can perform up to 65536 GFP_ATOMIC
allocations without coming up for air.  The possibility of ENOMEM or of
causing collateral problems is significant.  It doesn't look too hard
to change this to use GFP_KERNEL?

> +u64 kcov_common_handle(void)
> +{
> +	return current->kcov_handle;
> +}

I don't immediately understand what this "common handle" thing is all about. 
Code is rather lacking in this sort of high-level commentary?

Re: [PATCH v2 1/3] kcov: remote coverage support

[Dmitry Vyukov]

On Wed, Oct 23, 2019 at 5:24 PM Andrey Konovalov <andreyknvl@google.com> wrote:
>
> This patch adds background thread coverage collection ability to kcov.
...
> +static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
> +{
> +       struct kcov_remote *remote;
> +
> +       if (kcov_remote_find(handle))
> +               return ERR_PTR(-EEXIST);
> +       remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
> +       if (!remote)
> +               return ERR_PTR(-ENOMEM);
> +       remote->handle = handle;
> +       remote->kcov = kcov;
> +       hash_add(kcov_remote_map, &remote->hnode, handle);

I think it will make sense to check that there is no existing kcov
with the same handle registered. Such condition will be extremely hard
to debug based on episodically missing coverage.

...
>  void kcov_task_exit(struct task_struct *t)
>  {
>         struct kcov *kcov;
> @@ -256,15 +401,23 @@ void kcov_task_exit(struct task_struct *t)
>         kcov = t->kcov;
>         if (kcov == NULL)
>                 return;
> +
>         spin_lock(&kcov->lock);
> +       kcov_debug("t = %px, kcov->t = %px\n", t, kcov->t);
> +       /*
> +        * If !kcov->remote, this checks that t->kcov->t == t.
> +        * If kcov->remote == true then the exiting task is either:
> +        * 1. a remote task between kcov_remote_start() and kcov_remote_stop(),
> +        *    in this case t != kcov->t and we'll print a warning; or

Why? Is kcov->t == NULL for remote kcov's? May be worth mentioning in
the comment b/c it's a very condensed form to check lots of different
things at once.

Otherwise the series look good to me:

Reviewed-by: Dmitry Vyukov <dvyukov@google.com>

But Andrew's comments stand. It's possible I understand all of this
only because I already know how it works and why it works this way.

Re: [PATCH v2 0/3] kcov: collect coverage from usb and vhost

[Andrey Konovalov]

On Thu, Oct 24, 2019 at 12:04 AM Andrew Morton
<akpm@linux-foundation.org> wrote:
>
> On Wed, 23 Oct 2019 17:24:28 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:
>
> > This patchset extends kcov to allow collecting coverage from the USB
> > subsystem and vhost workers. See the first patch description for details
> > about the kcov extension. The other two patches apply this kcov extension
> > to USB and vhost.
> >
> > These patches have been used to enable coverage-guided USB fuzzing with
> > syzkaller for the last few years
>
> I find it surprising that this material is so focused on USB.  Is
> there something unique about USB that gave rise to this situation, or

USB fuzzing is the thing that I've been working on that requires this
functionality. But the idea is to make the interface generic enough to
make it useful for other subsystems as well as long as the annotations
are added.

> is it expected that the new kcov feature will be used elsewhere in the
> kernel?
>
> If the latter, which are the expected subsystems?

Currently we encountered two cases where this is useful: USB and vhost
workers. Most probably there are more subsystems that will benefit
from this kcov extension to get better fuzzing coverage. I don't have
a list of them, but the provided interface should be easy to use when
more of such cases are encountered.

Re: [PATCH v2 1/3] kcov: remote coverage support

[Andrey Konovalov]

On Thu, Oct 24, 2019 at 12:22 AM Andrew Morton
<akpm@linux-foundation.org> wrote:
>
> On Wed, 23 Oct 2019 17:24:29 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:
>
> > This patch adds background thread coverage collection ability to kcov.
> >
> > With KCOV_ENABLE coverage is collected only for syscalls that are issued
> > from the current process. With KCOV_REMOTE_ENABLE it's possible to collect
> > coverage for arbitrary parts of the kernel code, provided that those parts
> > are annotated with kcov_remote_start()/kcov_remote_stop().
> >
> > This allows to collect coverage from two types of kernel background
> > threads: the global ones, that are spawned during kernel boot and are
> > always running (e.g. USB hub_event()); and the local ones, that are
> > spawned when a user interacts with some kernel interface (e.g. vhost
> > workers).
> >
> > To enable collecting coverage from a global background thread, a unique
> > global handle must be assigned and passed to the corresponding
> > kcov_remote_start() call. Then a userspace process can pass a list of such
> > handles to the KCOV_REMOTE_ENABLE ioctl in the handles array field of the
> > kcov_remote_arg struct. This will attach the used kcov device to the code
> > sections, that are referenced by those handles.
> >
> > Since there might be many local background threads spawned from different
> > userspace processes, we can't use a single global handle per annotation.
> > Instead, the userspace process passes a non-zero handle through the
> > common_handle field of the kcov_remote_arg struct. This common handle gets
> > saved to the kcov_handle field in the current task_struct and needs to be
> > passed to the newly spawned threads via custom annotations. Those threads
> > should in turn be annotated with kcov_remote_start()/kcov_remote_stop().
> >
> > Internally kcov stores handles as u64 integers. The top byte of a handle
> > is used to denote the id of a subsystem that this handle belongs to, and
> > the lower 4 bytes are used to denote a handle id within that subsystem.
> > A reserved value 0 is used as a subsystem id for common handles as they
> > don't belong to a particular subsystem. The bytes 4-7 are currently
> > reserved and must be zero. In the future the number of bytes used for the
> > subsystem or handle ids might be increased.
> >
> > When a particular userspace proccess collects coverage by via a common
> > handle, kcov will collect coverage for each code section that is annotated
> > to use the common handle obtained as kcov_handle from the current
> > task_struct. However non common handles allow to collect coverage
> > selectively from different subsystems.
> >
> > ...
> >
> > +static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
> > +{
> > +     struct kcov_remote *remote;
> > +
> > +     if (kcov_remote_find(handle))
> > +             return ERR_PTR(-EEXIST);
> > +     remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
> > +     if (!remote)
> > +             return ERR_PTR(-ENOMEM);
> > +     remote->handle = handle;
> > +     remote->kcov = kcov;
> > +     hash_add(kcov_remote_map, &remote->hnode, handle);
> > +     return remote;
> > +}
> > +
> >
> > ...
> >
> > +             spin_lock(&kcov_remote_lock);
> > +             for (i = 0; i < remote_arg->num_handles; i++) {
> > +                     kcov_debug("handle %llx\n", remote_arg->handles[i]);
> > +                     if (!kcov_check_handle(remote_arg->handles[i],
> > +                                             false, true, false)) {
> > +                             spin_unlock(&kcov_remote_lock);
> > +                             kcov_disable(t, kcov);
> > +                             return -EINVAL;
> > +                     }
> > +                     remote = kcov_remote_add(kcov, remote_arg->handles[i]);
> > +                     if (IS_ERR(remote)) {
> > +                             spin_unlock(&kcov_remote_lock);
> > +                             kcov_disable(t, kcov);
> > +                             return PTR_ERR(remote);
> > +                     }
> > +             }
>
> It's worrisome that this code can perform up to 65536 GFP_ATOMIC
> allocations without coming up for air.  The possibility of ENOMEM or of
> causing collateral problems is significant.  It doesn't look too hard
> to change this to use GFP_KERNEL?

Sure, I'll do that in v3. I can also change the limit on the number of
handles to something lower (0x100?).

>
> > +u64 kcov_common_handle(void)
> > +{
> > +     return current->kcov_handle;
> > +}
>
> I don't immediately understand what this "common handle" thing is all about.
> Code is rather lacking in this sort of high-level commentary?

It's described in the documentation, but I'll add comments to the
exported functions to explain how they work and how they should be
used.

Thanks!

Re: [PATCH v2 1/3] kcov: remote coverage support

[Andrey Konovalov]

On Thu, Oct 24, 2019 at 9:27 AM Dmitry Vyukov <dvyukov@google.com> wrote:
>
> On Wed, Oct 23, 2019 at 5:24 PM Andrey Konovalov <andreyknvl@google.com> wrote:
> >
> > This patch adds background thread coverage collection ability to kcov.
> ...
> > +static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
> > +{
> > +       struct kcov_remote *remote;
> > +
> > +       if (kcov_remote_find(handle))
> > +               return ERR_PTR(-EEXIST);
> > +       remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
> > +       if (!remote)
> > +               return ERR_PTR(-ENOMEM);
> > +       remote->handle = handle;
> > +       remote->kcov = kcov;
> > +       hash_add(kcov_remote_map, &remote->hnode, handle);
>
> I think it will make sense to check that there is no existing kcov
> with the same handle registered. Such condition will be extremely hard
> to debug based on episodically missing coverage.

Will do in v3.

>
> ...
> >  void kcov_task_exit(struct task_struct *t)
> >  {
> >         struct kcov *kcov;
> > @@ -256,15 +401,23 @@ void kcov_task_exit(struct task_struct *t)
> >         kcov = t->kcov;
> >         if (kcov == NULL)
> >                 return;
> > +
> >         spin_lock(&kcov->lock);
> > +       kcov_debug("t = %px, kcov->t = %px\n", t, kcov->t);
> > +       /*
> > +        * If !kcov->remote, this checks that t->kcov->t == t.
> > +        * If kcov->remote == true then the exiting task is either:
> > +        * 1. a remote task between kcov_remote_start() and kcov_remote_stop(),
> > +        *    in this case t != kcov->t and we'll print a warning; or
>
> Why? Is kcov->t == NULL for remote kcov's? May be worth mentioning in
> the comment b/c it's a very condensed form to check lots of different
> things at once.

For remote kcov instances kcov->t points to the thread that created
the kcov device (I'll update the comment in struct kcov). When a task
is between kcov_remote_start() and kcov_remote_stop(), it's t->kcov
point to the remote kcov. So t is current for this task, and
t->kcov->t is the task that created the kcov instance. I'll expand the
comment to explain this better.

> Otherwise the series look good to me:
>
> Reviewed-by: Dmitry Vyukov <dvyukov@google.com>

Great, thanks!

>
> But Andrew's comments stand. It's possible I understand all of this
> only because I already know how it works and why it works this way.

Re: [PATCH v2 0/3] kcov: collect coverage from usb and vhost

[Andrew Morton]

On Thu, 24 Oct 2019 14:47:31 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:

> > is it expected that the new kcov feature will be used elsewhere in the
> > kernel?
> >
> > If the latter, which are the expected subsystems?
> 
> Currently we encountered two cases where this is useful: USB and vhost
> workers. Most probably there are more subsystems that will benefit
> from this kcov extension to get better fuzzing coverage. I don't have
> a list of them, but the provided interface should be easy to use when
> more of such cases are encountered.

It would be helpful to add such a list to the changelog.  Best-effort
and approximate is OK - just to help people understand the eventual
usefulness of the proposal.

Re: [PATCH v2 0/3] kcov: collect coverage from usb and vhost

[Andrey Konovalov]

On Fri, Oct 25, 2019 at 1:59 AM Andrew Morton <akpm@linux-foundation.org> wrote:
>
> On Thu, 24 Oct 2019 14:47:31 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:
>
> > > is it expected that the new kcov feature will be used elsewhere in the
> > > kernel?
> > >
> > > If the latter, which are the expected subsystems?
> >
> > Currently we encountered two cases where this is useful: USB and vhost
> > workers. Most probably there are more subsystems that will benefit
> > from this kcov extension to get better fuzzing coverage. I don't have
> > a list of them, but the provided interface should be easy to use when
> > more of such cases are encountered.
>
> It would be helpful to add such a list to the changelog.  Best-effort
> and approximate is OK - just to help people understand the eventual
> usefulness of the proposal.

OK, I'll add it to the cover letter in v3, thanks!

Re: [PATCH v2 3/3] vhost, kcov: collect coverage from vhost_worker

[kbuild test robot]

[-- Attachment #1: Type: text/plain, Size: 1598 bytes --]

Hi Andrey,

Thank you for the patch! Yet something to improve:

[auto build test ERROR on linus/master]
[cannot apply to v5.4-rc4 next-20191025]
[if your patch is applied to the wrong git tree, please drop us a note to help
improve the system. BTW, we also suggest to use '--base' option to specify the
base tree in git format-patch, please see https://stackoverflow.com/a/37406982]

url:    https://github.com/0day-ci/linux/commits/Andrey-Konovalov/kcov-collect-coverage-from-usb-and-vhost/20191025-145734
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git 39a38bcba4ab6e5285b07675b0e42c96eec35e67
config: s390-allmodconfig (attached as .config)
compiler: s390-linux-gcc (GCC) 7.4.0
reproduce:
        wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross
        chmod +x ~/bin/make.cross
        # save the attached .config to linux build tree
        GCC_VERSION=7.4.0 make.cross ARCH=s390 

If you fix the issue, kindly add following tag
Reported-by: kbuild test robot <lkp@intel.com>

All errors (new ones prefixed by >>):

>> ERROR: "kcov_common_handle" [drivers/vhost/vhost.ko] undefined!
   ERROR: "kcov_remote_stop" [drivers/vhost/vhost.ko] undefined!
   ERROR: "kcov_remote_start" [drivers/vhost/vhost.ko] undefined!
   ERROR: "kcov_remote_stop" [drivers/usb/core/usbcore.ko] undefined!
   ERROR: "kcov_remote_start" [drivers/usb/core/usbcore.ko] undefined!

---
0-DAY kernel test infrastructure                Open Source Technology Center
https://lists.01.org/pipermail/kbuild-all                   Intel Corporation

[-- Attachment #2: .config.gz --]
[-- Type: application/gzip, Size: 56369 bytes --]

Re: [PATCH v2 1/3] kcov: remote coverage support

[Andrey Konovalov]

On Thu, Oct 24, 2019 at 9:27 AM Dmitry Vyukov <dvyukov@google.com> wrote:
>
> On Wed, Oct 23, 2019 at 5:24 PM Andrey Konovalov <andreyknvl@google.com> wrote:
> >
> > This patch adds background thread coverage collection ability to kcov.
> ...
> > +static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
> > +{
> > +       struct kcov_remote *remote;
> > +
> > +       if (kcov_remote_find(handle))
> > +               return ERR_PTR(-EEXIST);
> > +       remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
> > +       if (!remote)
> > +               return ERR_PTR(-ENOMEM);
> > +       remote->handle = handle;
> > +       remote->kcov = kcov;
> > +       hash_add(kcov_remote_map, &remote->hnode, handle);
>
> I think it will make sense to check that there is no existing kcov
> with the same handle registered. Such condition will be extremely hard
> to debug based on episodically missing coverage.

Although looking at this again: we already check that by calling
kcov_remote_find().

>
> ...
> >  void kcov_task_exit(struct task_struct *t)
> >  {
> >         struct kcov *kcov;
> > @@ -256,15 +401,23 @@ void kcov_task_exit(struct task_struct *t)
> >         kcov = t->kcov;
> >         if (kcov == NULL)
> >                 return;
> > +
> >         spin_lock(&kcov->lock);
> > +       kcov_debug("t = %px, kcov->t = %px\n", t, kcov->t);
> > +       /*
> > +        * If !kcov->remote, this checks that t->kcov->t == t.
> > +        * If kcov->remote == true then the exiting task is either:
> > +        * 1. a remote task between kcov_remote_start() and kcov_remote_stop(),
> > +        *    in this case t != kcov->t and we'll print a warning; or
>
> Why? Is kcov->t == NULL for remote kcov's? May be worth mentioning in
> the comment b/c it's a very condensed form to check lots of different
> things at once.
>
> Otherwise the series look good to me:
>
> Reviewed-by: Dmitry Vyukov <dvyukov@google.com>
>
> But Andrew's comments stand. It's possible I understand all of this
> only because I already know how it works and why it works this way.

Re: [PATCH v2 1/3] kcov: remote coverage support

[Andrey Konovalov]

On Thu, Oct 24, 2019 at 12:22 AM Andrew Morton
<akpm@linux-foundation.org> wrote:
>
> On Wed, 23 Oct 2019 17:24:29 +0200 Andrey Konovalov <andreyknvl@google.com> wrote:
>
> > This patch adds background thread coverage collection ability to kcov.
> >
> > With KCOV_ENABLE coverage is collected only for syscalls that are issued
> > from the current process. With KCOV_REMOTE_ENABLE it's possible to collect
> > coverage for arbitrary parts of the kernel code, provided that those parts
> > are annotated with kcov_remote_start()/kcov_remote_stop().
> >
> > This allows to collect coverage from two types of kernel background
> > threads: the global ones, that are spawned during kernel boot and are
> > always running (e.g. USB hub_event()); and the local ones, that are
> > spawned when a user interacts with some kernel interface (e.g. vhost
> > workers).
> >
> > To enable collecting coverage from a global background thread, a unique
> > global handle must be assigned and passed to the corresponding
> > kcov_remote_start() call. Then a userspace process can pass a list of such
> > handles to the KCOV_REMOTE_ENABLE ioctl in the handles array field of the
> > kcov_remote_arg struct. This will attach the used kcov device to the code
> > sections, that are referenced by those handles.
> >
> > Since there might be many local background threads spawned from different
> > userspace processes, we can't use a single global handle per annotation.
> > Instead, the userspace process passes a non-zero handle through the
> > common_handle field of the kcov_remote_arg struct. This common handle gets
> > saved to the kcov_handle field in the current task_struct and needs to be
> > passed to the newly spawned threads via custom annotations. Those threads
> > should in turn be annotated with kcov_remote_start()/kcov_remote_stop().
> >
> > Internally kcov stores handles as u64 integers. The top byte of a handle
> > is used to denote the id of a subsystem that this handle belongs to, and
> > the lower 4 bytes are used to denote a handle id within that subsystem.
> > A reserved value 0 is used as a subsystem id for common handles as they
> > don't belong to a particular subsystem. The bytes 4-7 are currently
> > reserved and must be zero. In the future the number of bytes used for the
> > subsystem or handle ids might be increased.
> >
> > When a particular userspace proccess collects coverage by via a common
> > handle, kcov will collect coverage for each code section that is annotated
> > to use the common handle obtained as kcov_handle from the current
> > task_struct. However non common handles allow to collect coverage
> > selectively from different subsystems.
> >
> > ...
> >
> > +static struct kcov_remote *kcov_remote_add(struct kcov *kcov, u64 handle)
> > +{
> > +     struct kcov_remote *remote;
> > +
> > +     if (kcov_remote_find(handle))
> > +             return ERR_PTR(-EEXIST);
> > +     remote = kmalloc(sizeof(*remote), GFP_ATOMIC);
> > +     if (!remote)
> > +             return ERR_PTR(-ENOMEM);
> > +     remote->handle = handle;
> > +     remote->kcov = kcov;
> > +     hash_add(kcov_remote_map, &remote->hnode, handle);
> > +     return remote;
> > +}
> > +
> >
> > ...
> >
> > +             spin_lock(&kcov_remote_lock);
> > +             for (i = 0; i < remote_arg->num_handles; i++) {
> > +                     kcov_debug("handle %llx\n", remote_arg->handles[i]);
> > +                     if (!kcov_check_handle(remote_arg->handles[i],
> > +                                             false, true, false)) {
> > +                             spin_unlock(&kcov_remote_lock);
> > +                             kcov_disable(t, kcov);
> > +                             return -EINVAL;
> > +                     }
> > +                     remote = kcov_remote_add(kcov, remote_arg->handles[i]);
> > +                     if (IS_ERR(remote)) {
> > +                             spin_unlock(&kcov_remote_lock);
> > +                             kcov_disable(t, kcov);
> > +                             return PTR_ERR(remote);
> > +                     }
> > +             }
>
> It's worrisome that this code can perform up to 65536 GFP_ATOMIC
> allocations without coming up for air.  The possibility of ENOMEM or of
> causing collateral problems is significant.  It doesn't look too hard
> to change this to use GFP_KERNEL?

Looking at this again: it seems easy to get rid of locking
kcov_remote_lock when doing kmalloc, but a bit harder to get rid of
kcov->lock. Andrew, would it be OK to just change the max number of
GFP_ATOMIC allocations to 256?

>
> > +u64 kcov_common_handle(void)
> > +{
> > +     return current->kcov_handle;
> > +}
>
> I don't immediately understand what this "common handle" thing is all about.
> Code is rather lacking in this sort of high-level commentary?
>
>