summaryrefslogtreecommitdiffstats
path: root/drivers/nvme/host/ioctl.c
diff options
context:
space:
mode:
Diffstat (limited to 'drivers/nvme/host/ioctl.c')
-rw-r--r--drivers/nvme/host/ioctl.c138
1 files changed, 87 insertions, 51 deletions
diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c
index 9ddda571f046..06f52db34be9 100644
--- a/drivers/nvme/host/ioctl.c
+++ b/drivers/nvme/host/ioctl.c
@@ -8,13 +8,27 @@
#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,
- fmode_t mode)
+ 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.
*/
@@ -43,11 +57,29 @@ static bool nvme_cmd_allowed(struct nvme_ns *ns, struct nvme_command *c,
}
/*
- * Only allow I/O commands that transfer data to the controller if the
- * special file is open for writing, but always allow I/O commands that
- * transfer data from the controller.
+ * 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))
+ if (nvme_is_write(c) || (effects & NVME_CMD_EFFECTS_LBCC))
return mode & FMODE_WRITE;
return true;
}
@@ -130,7 +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;
@@ -143,7 +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);
@@ -152,8 +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)
@@ -183,9 +215,9 @@ 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_ctrl *ctrl;
struct request *req;
@@ -201,7 +233,7 @@ 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;
}
@@ -284,10 +316,8 @@ 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,
@@ -305,7 +335,8 @@ 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, fmode_t mode)
+ struct nvme_passthru_cmd __user *ucmd, unsigned int flags,
+ fmode_t mode)
{
struct nvme_passthru_cmd cmd;
struct nvme_command c;
@@ -333,16 +364,15 @@ 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, mode))
+ 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))
@@ -353,8 +383,8 @@ 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,
- fmode_t mode)
+ struct nvme_passthru_cmd64 __user *ucmd, unsigned int flags,
+ fmode_t mode)
{
struct nvme_passthru_cmd64 cmd;
struct nvme_command c;
@@ -381,16 +411,15 @@ 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, mode))
+ 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))
@@ -551,7 +580,7 @@ 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, ioucmd->file->f_mode))
+ if (!nvme_cmd_allowed(ns, &c, 0, ioucmd->file->f_mode))
return -EACCES;
d.metadata = READ_ONCE(cmd->metadata);
@@ -621,9 +650,9 @@ static int nvme_ctrl_ioctl(struct nvme_ctrl *ctrl, unsigned int cmd,
{
switch (cmd) {
case NVME_IOCTL_ADMIN_CMD:
- return nvme_user_cmd(ctrl, NULL, argp, mode);
+ return nvme_user_cmd(ctrl, NULL, argp, 0, mode);
case NVME_IOCTL_ADMIN64_CMD:
- return nvme_user_cmd64(ctrl, NULL, argp, false, mode);
+ return nvme_user_cmd64(ctrl, NULL, argp, 0, mode);
default:
return sed_ioctl(ctrl->opal_dev, cmd, argp);
}
@@ -648,14 +677,14 @@ struct nvme_user_io32 {
#endif /* COMPAT_FOR_U64_ALIGNMENT */
static int nvme_ns_ioctl(struct nvme_ns *ns, unsigned int cmd,
- void __user *argp, fmode_t mode)
+ 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, mode);
+ 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
@@ -666,37 +695,40 @@ 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, mode);
case NVME_IOCTL_IO64_CMD_VEC:
- return nvme_user_cmd64(ns->ctrl, ns, argp, true, mode);
+ 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,
- fmode_t mode)
-{
- if (is_ctrl_ioctl(cmd))
- return nvme_ctrl_ioctl(ns->ctrl, cmd, arg, mode);
- return nvme_ns_ioctl(ns, cmd, arg, mode);
-}
-
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;
+
+ if (bdev_is_partition(bdev))
+ flags |= NVME_IOCTL_PARTITION;
- return __nvme_ioctl(ns, cmd, (void __user *)arg, mode);
+ 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, file->f_mode);
+ 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)
@@ -786,6 +818,10 @@ 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);
@@ -801,7 +837,7 @@ int nvme_ns_head_ioctl(struct block_device *bdev, fmode_t mode,
return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx,
mode);
- ret = nvme_ns_ioctl(ns, cmd, argp, mode);
+ ret = nvme_ns_ioctl(ns, cmd, argp, flags, mode);
out_unlock:
srcu_read_unlock(&head->srcu, srcu_idx);
return ret;
@@ -826,7 +862,7 @@ long nvme_ns_head_chr_ioctl(struct file *file, unsigned int cmd,
return nvme_ns_head_ctrl_ioctl(ns, cmd, argp, head, srcu_idx,
file->f_mode);
- ret = nvme_ns_ioctl(ns, cmd, argp, file->f_mode);
+ ret = nvme_ns_ioctl(ns, cmd, argp, 0, file->f_mode);
out_unlock:
srcu_read_unlock(&head->srcu, srcu_idx);
return ret;
@@ -925,7 +961,7 @@ 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, mode);
+ ret = nvme_user_cmd(ctrl, ns, argp, 0, mode);
nvme_put_ns(ns);
return ret;
@@ -942,9 +978,9 @@ long nvme_dev_ioctl(struct file *file, unsigned int cmd,
switch (cmd) {
case NVME_IOCTL_ADMIN_CMD:
- return nvme_user_cmd(ctrl, NULL, argp, file->f_mode);
+ return nvme_user_cmd(ctrl, NULL, argp, 0, file->f_mode);
case NVME_IOCTL_ADMIN64_CMD:
- return nvme_user_cmd64(ctrl, NULL, argp, false, file->f_mode);
+ return nvme_user_cmd64(ctrl, NULL, argp, 0, file->f_mode);
case NVME_IOCTL_IO_CMD:
return nvme_dev_user_cmd(ctrl, argp, file->f_mode);
case NVME_IOCTL_RESET:
generated by cgit v1.2.3 (git 2.39.1) at 2024-05-23 17:26:36 +0000