1 files changed, 148 insertions, 59 deletions

diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c
index 81f5550b670d..723e7d5b778f 100644
--- a/drivers/nvme/host/ioctl.c
+++ b/drivers/nvme/host/ioctl.c
@@ -8,6 +8,82 @@
 #include <linux/io_uring.h>
 #include "nvme.h"
 
+enum {
+	NVME_IOCTL_VEC		= (1 << 0),
+	NVME_IOCTL_PARTITION	= (1 << 1),
+};
+
+static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c,
+		unsigned int flags, fmode_t mode)
+{
+	u32 effects;
+
+	if (capable(CAP_SYS_ADMIN))
+		return true;
+
+	/*
+	 * Do not allow unprivileged passthrough on partitions, as that allows an
+	 * escape from the containment of the partition.
+	 */
+	if (flags & NVME_IOCTL_PARTITION)
+		return false;
+
+	/*
+	 * Do not allow unprivileged processes to send vendor specific or fabrics
+	 * commands as we can't be sure about their effects.
+	 */
+	if (c->common.opcode >= nvme_cmd_vendor_start ||
+	    c->common.opcode == nvme_fabrics_command)
+		return false;
+
+	/*
+	 * Do not allow unprivileged passthrough of admin commands except
+	 * for a subset of identify commands that contain information required
+	 * to form proper I/O commands in userspace and do not expose any
+	 * potentially sensitive information.
+	 */
+	if (!ns) {
+		if (c->common.opcode == nvme_admin_identify) {
+			switch (c->identify.cns) {
+			case NVME_ID_CNS_NS:
+			case NVME_ID_CNS_CS_NS:
+			case NVME_ID_CNS_NS_CS_INDEP:
+			case NVME_ID_CNS_CS_CTRL:
+			case NVME_ID_CNS_CTRL:
+				return true;
+			}
+		}
+		return false;
+	}
+
+	/*
+	 * Check if the controller provides a Commands Supported and Effects log
+	 * and marks this command as supported.  If not reject unprivileged
+	 * passthrough.
+	 */
+	effects = nvme_command_effects(ns->ctrl, ns, c->common.opcode);
+	if (!(effects & NVME_CMD_EFFECTS_CSUPP))
+		return false;
+
+	/*
+	 * Don't allow passthrough for command that have intrusive (or unknown)
+	 * effects.
+	 */
+	if (effects & ~(NVME_CMD_EFFECTS_CSUPP | NVME_CMD_EFFECTS_LBCC |
+			NVME_CMD_EFFECTS_UUID_SEL |
+			NVME_CMD_EFFECTS_SCOPE_MASK))
+		return false;
+
+	/*
+	 * Only allow I/O commands that transfer data to the controller or that
+	 * change the logical block contents if the file descriptor is open for
+	 * writing.
+	 */
+	if (nvme_is_write(c) || (effects & NVME_CMD_EFFECTS_LBCC))
+		return mode & FMODE_WRITE;
+	return true;
+}
+
 /*
  * Convert integer values from ioctl structures to user pointers, silently
  * ignoring the upper bits in the compat case to match behaviour of 32-bit
@@ -86,7 +162,7 @@ static struct request *nvme_alloc_user_request(struct request_queue *q,
 static int nvme_map_user_request(struct request *req, u64 ubuffer,
 		unsigned bufflen, void __user *meta_buffer, unsigned meta_len,
 		u32 meta_seed, void **metap, struct io_uring_cmd *ioucmd,
-		bool vec)
+		unsigned int flags)
 {
 	struct request_queue *q = req->q;
 	struct nvme_ns *ns = q->queuedata;
@@ -99,7 +175,7 @@ static int nvme_map_user_request(struct request *req, u64 ubuffer,
 		struct iov_iter iter;
 
 		/* fixedbufs is only for non-vectored io */
-		if (WARN_ON_ONCE(vec))
+		if (WARN_ON_ONCE(flags & NVME_IOCTL_VEC))
 			return -EINVAL;
 		ret = io_uring_cmd_import_fixed(ubuffer, bufflen,
 				rq_data_dir(req), &iter, ioucmd);
@@ -108,8 +184,8 @@ static int nvme_map_user_request(struct request *req, u64 ubuffer,
 		ret = blk_rq_map_user_iov(q, req, NULL, &iter, GFP_KERNEL);
 	} else {
 		ret = blk_rq_map_user_io(req, NULL, nvme_to_user_ptr(ubuffer),
-				bufflen, GFP_KERNEL, vec, 0, 0,
-				rq_data_dir(req));
+				bufflen, GFP_KERNEL, flags & NVME_IOCTL_VEC, 0,
+				0, rq_data_dir(req));
 	}
 
 	if (ret)
@@ -139,10 +215,11 @@ out:
 }
 
 static int nvme_submit_user_cmd(struct request_queue *q,
-		struct nvme_command *cmd, u64 ubuffer,
-		unsigned bufflen, void __user *meta_buffer, unsigned meta_len,
-		u32 meta_seed, u64 *result, unsigned timeout, bool vec)
+		struct nvme_command *cmd, u64 ubuffer, unsigned bufflen,
+		void __user *meta_buffer, unsigned meta_len, u32 meta_seed,
+		u64 *result, unsigned timeout, unsigned int flags)
 {
+	struct nvme_ns *ns = q->queuedata;
 	struct nvme_ctrl *ctrl;
 	struct request *req;
 	void *meta = NULL;
@@ -157,7 +234,7 @@ static int nvme_submit_user_cmd(struct request_queue *q,
 	req->timeout = timeout;
 	if (ubuffer && bufflen) {
 		ret = nvme_map_user_request(req, ubuffer, bufflen, meta_buffer,
-				meta_len, meta_seed, &meta, NULL, vec);
+				meta_len, meta_seed, &meta, NULL, flags);
 		if (ret)
 			return ret;
 	}
@@ -165,8 +242,8 @@ static int nvme_submit_user_cmd(struct request_queue *q,
 	bio = req->bio;
 	ctrl = nvme_req(req)->ctrl;
 
-	ret = nvme_execute_passthru_rq(req, &effects);
-
+	effects = nvme_passthru_start(ctrl, ns, cmd->common.opcode);
+	ret = nvme_execute_rq(req, false);
 	if (result)
 		*result = le64_to_cpu(nvme_req(req)->result.u64);
 	if (meta)
@@ -240,10 +317,8 @@ static int nvme_submit_io(struct nvme_ns *ns, struct nvme_user_io __user *uio)
 	c.rw.apptag = cpu_to_le16(io.apptag);
 	c.rw.appmask = cpu_to_le16(io.appmask);
 
-	return nvme_submit_user_cmd(ns->queue, &c,
-			io.addr, length,
-			metadata, meta_len, lower_32_bits(io.slba), NULL, 0,
-			false);
+	return nvme_submit_user_cmd(ns->queue, &c, io.addr, length, metadata,
+			meta_len, lower_32_bits(io.slba), NULL, 0, 0);
 }
 
 static bool nvme_validate_passthru_nsid(struct nvme_ctrl *ctrl,
@@ -261,7 +336,8 @@ static bool nvme_validate_passthru_nsid(struct nvme_ctrl *ctrl,
 }
 
 static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
-			struct nvme_passthru_cmd __user *ucmd)
+		struct nvme_passthru_cmd __user *ucmd, unsigned int flags,
+		fmode_t mode)
 {
 	struct nvme_passthru_cmd cmd;
 	struct nvme_command c;
@@ -269,8 +345,6 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	u64 result;
 	int status;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
 	if (copy_from_user(&cmd, ucmd, sizeof(cmd)))
 		return -EFAULT;
 	if (cmd.flags)
@@ -291,13 +365,15 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	c.common.cdw14 = cpu_to_le32(cmd.cdw14);
 	c.common.cdw15 = cpu_to_le32(cmd.cdw15);
 
+	if (!nvme_cmd_allowed(ns, &c, 0, mode))
+		return -EACCES;
+
 	if (cmd.timeout_ms)
 		timeout = msecs_to_jiffies(cmd.timeout_ms);
 
 	status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c,
-			cmd.addr, cmd.data_len,
-			nvme_to_user_ptr(cmd.metadata), cmd.metadata_len,
-			0, &result, timeout, false);
+			cmd.addr, cmd.data_len, nvme_to_user_ptr(cmd.metadata),
+			cmd.metadata_len, 0, &result, timeout, 0);
 
 	if (status >= 0) {
 		if (put_user(result, &ucmd->result))
@@ -308,15 +384,14 @@ static int nvme_user_cmd(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 }
 
 static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
-			struct nvme_passthru_cmd64 __user *ucmd, bool vec)
+		struct nvme_passthru_cmd64 __user *ucmd, unsigned int flags,
+		fmode_t mode)
 {
 	struct nvme_passthru_cmd64 cmd;
 	struct nvme_command c;
 	unsigned timeout = 0;
 	int status;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
 	if (copy_from_user(&cmd, ucmd, sizeof(cmd)))
 		return -EFAULT;
 	if (cmd.flags)
@@ -337,13 +412,15 @@ static int nvme_user_cmd64(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	c.common.cdw14 = cpu_to_le32(cmd.cdw14);
 	c.common.cdw15 = cpu_to_le32(cmd.cdw15);
 
+	if (!nvme_cmd_allowed(ns, &c, flags, mode))
+		return -EACCES;
+
 	if (cmd.timeout_ms)
 		timeout = msecs_to_jiffies(cmd.timeout_ms);
 
 	status = nvme_submit_user_cmd(ns ? ns->queue : ctrl->admin_q, &c,
-			cmd.addr, cmd.data_len,
-			nvme_to_user_ptr(cmd.metadata), cmd.metadata_len,
-			0, &cmd.result, timeout, vec);
+			cmd.addr, cmd.data_len, nvme_to_user_ptr(cmd.metadata),
+			cmd.metadata_len, 0, &cmd.result, timeout, flags);
 
 	if (status >= 0) {
 		if (put_user(cmd.result, &ucmd->result))
@@ -478,14 +555,11 @@ static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	struct nvme_uring_data d;
 	struct nvme_command c;
 	struct request *req;
-	blk_opf_t rq_flags = 0;
+	blk_opf_t rq_flags = REQ_ALLOC_CACHE;
 	blk_mq_req_flags_t blk_flags = 0;
 	void *meta = NULL;
 	int ret;
 
-	if (!capable(CAP_SYS_ADMIN))
-		return -EACCES;
-
 	c.common.opcode = READ_ONCE(cmd->opcode);
 	c.common.flags = READ_ONCE(cmd->flags);
 	if (c.common.flags)
@@ -507,6 +581,9 @@ static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	c.common.cdw14 = cpu_to_le32(READ_ONCE(cmd->cdw14));
 	c.common.cdw15 = cpu_to_le32(READ_ONCE(cmd->cdw15));
 
+	if (!nvme_cmd_allowed(ns, &c, 0, ioucmd->file->f_mode))
+		return -EACCES;
+
 	d.metadata = READ_ONCE(cmd->metadata);
 	d.addr = READ_ONCE(cmd->addr);
 	d.data_len = READ_ONCE(cmd->data_len);
@@ -514,7 +591,7 @@ static int nvme_uring_cmd_io(struct nvme_ctrl *ctrl, struct nvme_ns *ns,
 	d.timeout_ms = READ_ONCE(cmd->timeout_ms);
 
 	if (issue_flags & IO_URING_F_NONBLOCK) {
-		rq_flags = REQ_NOWAIT;
+		rq_flags |= REQ_NOWAIT;
 		blk_flags = BLK_MQ_REQ_NOWAIT;
 	}
 	if (issue_flags & IO_URING_F_IOPOLL)
@@ -570,13 +647,13 @@ static bool is_ctrl_ioctl(unsigned int cmd)
 }
 
 static int nvme_ctrl_ioctl(struct nvme_ctrl *ctrl, unsigned int cmd,
-		void __user *argp)
+		void __user *argp, fmode_t mode)
 {
 	switch (cmd) {
 	case NVME_IOCTL_ADMIN_CMD:
-		return nvme_user_cmd(ctrl, NULL, argp);
+		return nvme_user_cmd(ctrl, NULL, argp, 0, mode);
 	case NVME_IOCTL_ADMIN64_CMD:
-		return nvme_user_cmd64(ctrl, NULL, argp, false);
+		return nvme_user_cmd64(ctrl, NULL, argp, 0, mode);
 	default:
 		return sed_ioctl(ctrl->opal_dev, cmd, argp);
 	}
@@ -601,14 +678,14 @@ struct nvme_user_io32 {
 #endif /* COMPAT_FOR_U64_ALIGNMENT */
 
 static int nvme_ns_ioctl(struct nvme_ns *ns, unsigned int cmd,
-		void __user *argp)
+		void __user *argp, unsigned int flags, fmode_t mode)
 {
 	switch (cmd) {
 	case NVME_IOCTL_ID:
 		force_successful_syscall_return();
 		return ns->head->ns_id;
 	case NVME_IOCTL_IO_CMD:
-		return nvme_user_cmd(ns->ctrl, ns, argp);
+		return nvme_user_cmd(ns->ctrl, ns, argp, flags, mode);
 	/*
 	 * struct nvme_user_io can have different padding on some 32-bit ABIs.
 	 * Just accept the compat version as all fields that are used are the
@@ -619,36 +696,40 @@ static int nvme_ns_ioctl(struct nvme_ns *ns, unsigned int cmd,
 #endif
 	case NVME_IOCTL_SUBMIT_IO:
 		return nvme_submit_io(ns, argp);
-	case NVME_IOCTL_IO64_CMD:
-		return nvme_user_cmd64(ns->ctrl, ns, argp, false);
 	case NVME_IOCTL_IO64_CMD_VEC:
-		return nvme_user_cmd64(ns->ctrl, ns, argp, true);
+		flags |= NVME_IOCTL_VEC;
+		fallthrough;
+	case NVME_IOCTL_IO64_CMD:
+		return nvme_user_cmd64(ns->ctrl, ns, argp, flags, mode);
 	default:
 		return -ENOTTY;
 	}
 }
 
-static int __nvme_ioctl(struct nvme_ns *ns, unsigned int cmd, void __user *arg)
-{
-       if (is_ctrl_ioctl(cmd))
-               return nvme_ctrl_ioctl(ns->ctrl, cmd, arg);
-       return nvme_ns_ioctl(ns, cmd, arg);
-}
-
 int nvme_ioctl(struct block_device *bdev, fmode_t mode,
 		unsigned int cmd, unsigned long arg)
 {
 	struct nvme_ns *ns = bdev->bd_disk->private_data;
+	void __user *argp = (void __user *)arg;
+	unsigned int flags = 0;
 
-	return __nvme_ioctl(ns, cmd, (void __user *)arg);
+	if (bdev_is_partition(bdev))
+		flags |= NVME_IOCTL_PARTITION;
+
+	if (is_ctrl_ioctl(cmd))
+		return nvme_ctrl_ioctl(ns->ctrl, cmd, argp, mode);
+	return nvme_ns_ioctl(ns, cmd, argp, flags, mode);
 }
 
 long nvme_ns_chr_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
 {
 	struct nvme_ns *ns =
 		container_of(file_inode(file)->i_cdev, struct nvme_ns, cdev);
+	void __user *argp = (void __user *)arg;
 
-	return __nvme_ioctl(ns, cmd, (void __user *)arg);
+	if (is_ctrl_ioctl(cmd))
+		return nvme_ctrl_ioctl(ns->ctrl, cmd, argp, file->f_mode);
+	return nvme_ns_ioctl(ns, cmd, argp, 0, file->f_mode);
 }
 
 static int nvme_uring_cmd_checks(unsigned int issue_flags)
@@ -716,7 +797,8 @@ int nvme_ns_chr_uring_cmd_iopoll(struct io_uring_cmd *ioucmd,
 }
 #ifdef CONFIG_NVME_MULTIPATH
 static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd,
-		void __user *argp, struct nvme_ns_head *head, int srcu_idx)
+		void __user *argp, struct nvme_ns_head *head, int srcu_idx,
+		fmode_t mode)
 	__releases(&head->srcu)
 {
 	struct nvme_ctrl *ctrl = ns->ctrl;
@@ -724,7 +806,7 @@ static int nvme_ns_head_ctrl_ioctl(struct nvme_ns *ns, unsigned int cmd,
 
 	nvme_get_ctrl(ns->ctrl);
 	srcu_read_unlock(&head->srcu, srcu_idx);
-	ret = nvme_ctrl_ioctl(ns->ctrl, cmd, argp);
+	ret = nvme_ctrl_ioctl(ns->ctrl, cmd, argp, mode);
 
 	nvme_put_ctrl(ctrl);
 	return ret;
@@ -737,6 +819,10 @@ int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode,
 	void __user *argp = (void __user *)arg;
 	struct nvme_ns *ns;
 	int srcu_idx, ret = -EWOULDBLOCK;
+	unsigned int flags = 0;
+
+	if (bdev_is_partition(bdev))
+		flags |= NVME_IOCTL_PARTITION;
 
 	srcu_idx = srcu_read_lock(&head->srcu);
 	ns = nvme_find_path(head);
@@ -749,9 +835,10 @@ int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode,
 	 * deadlock when deleting namespaces using the passthrough interface.
 	 */
 	if (is_ctrl_ioctl(cmd))
-		return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx);
+		return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx,
+					mode);
 
-	ret = nvme_ns_ioctl(ns, cmd, argp);
+	ret = nvme_ns_ioctl(ns, cmd, argp, flags, mode);
 out_unlock:
 	srcu_read_unlock(&head->srcu, srcu_idx);
 	return ret;
@@ -773,9 +860,10 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,
 		goto out_unlock;
 
 	if (is_ctrl_ioctl(cmd))
-		return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx);
+		return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx,
+				file->f_mode);
 
-	ret = nvme_ns_ioctl(ns, cmd, argp);
+	ret = nvme_ns_ioctl(ns, cmd, argp, 0, file->f_mode);
 out_unlock:
 	srcu_read_unlock(&head->srcu, srcu_idx);
 	return ret;
@@ -849,7 +937,8 @@ int nvme_dev_uring_cmd(struct io_uring_cmd *ioucmd, unsigned int issue_flags)
 	return ret;
 }
 
-static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
+static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp,
+		fmode_t mode)
 {
 	struct nvme_ns *ns;
 	int ret;
@@ -873,7 +962,7 @@ static int nvme_dev_user_cmd(struct nvme_ctrl *ctrl, void __user *argp)
 	kref_get(&ns->kref);
 	up_read(&ctrl->namespaces_rwsem);
 
-	ret = nvme_user_cmd(ctrl, ns, argp);
+	ret = nvme_user_cmd(ctrl, ns, argp, 0, mode);
 	nvme_put_ns(ns);
 	return ret;
 
@@ -890,11 +979,11 @@ long nvme_dev_ioctl(struct file *file, unsigned int cmd,
 
 	switch (cmd) {
 	case NVME_IOCTL_ADMIN_CMD:
-		return nvme_user_cmd(ctrl, NULL, argp);
+		return nvme_user_cmd(ctrl, NULL, argp, 0, file->f_mode);
 	case NVME_IOCTL_ADMIN64_CMD:
-		return nvme_user_cmd64(ctrl, NULL, argp, false);
+		return nvme_user_cmd64(ctrl, NULL, argp, 0, file->f_mode);
 	case NVME_IOCTL_IO_CMD:
-		return nvme_dev_user_cmd(ctrl, argp);
+		return nvme_dev_user_cmd(ctrl, argp, file->f_mode);
 	case NVME_IOCTL_RESET:
 		if (!capable(CAP_SYS_ADMIN))
 			return -EACCES;