Commit 3d25b1e8 authored by Jens Axboe's avatar Jens Axboe

Merge tag 'nvme-6.2-2023-01-12' of git://git.infradead.org/nvme into block-6.2

Pull NVMe fixes from Christoph:

"nvme fixes for Linux 6.2

 - Identify quirks for Apple controllers (Hector Martin)
 - fix error handling in nvme_pci_enable (Tong Zhang)
 - refuse unprivileged passthrough on partitions (Christoph Hellwig)
 - fix MAINTAINERS to not match nvmem subsystem headers (Russell King)"

* tag 'nvme-6.2-2023-01-12' of git://git.infradead.org/nvme:
  MAINTAINERS: stop nvme matching for nvmem files
  nvme: don't allow unprivileged passthrough on partitions
  nvme: replace the "bool vec" arguments with flags in the ioctl path
  nvme: remove __nvme_ioctl
  nvme-pci: fix error handling in nvme_pci_enable()
  nvme-pci: add NVME_QUIRK_IDENTIFY_CNS quirk to Apple T2 controllers
  nvme-apple: add NVME_QUIRK_IDENTIFY_CNS quirk to fix regression
parents 49e4d04f c7c0644e
......@@ -14830,7 +14830,8 @@ T: git://git.infradead.org/nvme.git
F: Documentation/nvme/
F: drivers/nvme/host/
F: drivers/nvme/common/
F: include/linux/nvme*
F: include/linux/nvme.h
F: include/linux/nvme-*.h
F: include/uapi/linux/nvme_ioctl.h
NVM EXPRESS FABRICS AUTHENTICATION
......
......@@ -1493,7 +1493,7 @@ static int apple_nvme_probe(struct platform_device *pdev)
}
ret = nvme_init_ctrl(&anv->ctrl, anv->dev, &nvme_ctrl_ops,
NVME_QUIRK_SKIP_CID_GEN);
NVME_QUIRK_SKIP_CID_GEN | NVME_QUIRK_IDENTIFY_CNS);
if (ret) {
dev_err_probe(dev, ret, "Failed to initialize nvme_ctrl");
goto put_dev;
......
......@@ -8,14 +8,26 @@
#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.
......@@ -150,7 +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;
......@@ -163,7 +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);
......@@ -172,8 +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)
......@@ -203,9 +215,9 @@ static int nvme_map_user_request(struct request *req, u64 ubuffer,
}
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;
......@@ -221,7 +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;
}
......@@ -304,10 +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,
......@@ -325,7 +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;
......@@ -353,16 +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))
......@@ -373,8 +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;
......@@ -401,16 +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))
......@@ -571,7 +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);
......@@ -641,9 +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);
}
......@@ -668,14 +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
......@@ -686,37 +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;
return __nvme_ioctl(ns, cmd, (void __user *)arg, mode);
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, 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)
......@@ -806,6 +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);
......@@ -821,7 +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;
......@@ -846,7 +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;
......@@ -945,7 +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;
......@@ -962,9 +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:
......
......@@ -2533,7 +2533,7 @@ static int nvme_pci_enable(struct nvme_dev *dev)
*/
result = pci_alloc_irq_vectors(pdev, 1, 1, PCI_IRQ_ALL_TYPES);
if (result < 0)
return result;
goto disable;
dev->ctrl.cap = lo_hi_readq(dev->bar + NVME_REG_CAP);
......@@ -2586,8 +2586,13 @@ static int nvme_pci_enable(struct nvme_dev *dev)
pci_enable_pcie_error_reporting(pdev);
pci_save_state(pdev);
return nvme_pci_configure_admin_queue(dev);
result = nvme_pci_configure_admin_queue(dev);
if (result)
goto free_irq;
return result;
free_irq:
pci_free_irq_vectors(pdev);
disable:
pci_disable_device(pdev);
return result;
......@@ -3495,7 +3500,8 @@ static const struct pci_device_id nvme_id_table[] = {
.driver_data = NVME_QUIRK_SINGLE_VECTOR |
NVME_QUIRK_128_BYTES_SQES |
NVME_QUIRK_SHARED_TAGS |
NVME_QUIRK_SKIP_CID_GEN },
NVME_QUIRK_SKIP_CID_GEN |
NVME_QUIRK_IDENTIFY_CNS },
{ PCI_DEVICE_CLASS(PCI_CLASS_STORAGE_EXPRESS, 0xffffff) },
{ 0, }
};
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment