diff mbox series

[v4,2/3] io_uring: add IOURING_REGISTER_RESTRICTIONS opcode

Message ID 20200813153254.93731-3-sgarzare@redhat.com (mailing list archive)
State New, archived
Headers show
Series io_uring: add restrictions to support untrusted applications and guests | expand

Commit Message

Stefano Garzarella Aug. 13, 2020, 3:32 p.m. UTC 
The new io_uring_register(2) IOURING_REGISTER_RESTRICTIONS opcode
permanently installs a feature allowlist on an io_ring_ctx.
The io_ring_ctx can then be passed to untrusted code with the
knowledge that only operations present in the allowlist can be
executed.

The allowlist approach ensures that new features added to io_uring
do not accidentally become available when an existing application
is launched on a newer kernel version.

Currently is it possible to restrict sqe opcodes, sqe flags, and
register opcodes.

IOURING_REGISTER_RESTRICTIONS can only be made once. Afterwards
it is not possible to change restrictions anymore.
This prevents untrusted code from removing restrictions.

Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---
v3:
 - added IORING_RESTRICTION_SQE_FLAGS_ALLOWED and
   IORING_RESTRICTION_SQE_FLAGS_REQUIRED
 - removed IORING_RESTRICTION_FIXED_FILES_ONLY

RFC v2:
 - added 'restricted' flag in the ctx [Jens]
 - added IORING_MAX_RESTRICTIONS define
 - returned EBUSY instead of EINVAL when restrictions are already
   registered
 - reset restrictions if an error happened during the registration
---
 fs/io_uring.c                 | 112 +++++++++++++++++++++++++++++++++-
 include/uapi/linux/io_uring.h |  31 ++++++++++
 2 files changed, 142 insertions(+), 1 deletion(-)

Comments

kernel test robot Aug. 13, 2020, 5:42 p.m. UTC | #1 
Hi Stefano,

I love your patch! Perhaps something to improve:

[auto build test WARNING on linus/master]
[also build test WARNING on v5.8 next-20200813]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch]

url:    https://github.com/0day-ci/linux/commits/Stefano-Garzarella/io_uring-add-restrictions-to-support-untrusted-applications-and-guests/20200813-233653
base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git dc06fe51d26efc100ac74121607c01a454867c91
config: s390-randconfig-c003-20200813 (attached as .config)
compiler: s390-linux-gcc (GCC) 9.3.0

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


coccinelle warnings: (new ones prefixed by >>)

>> fs/io_uring.c:8516:7-14: WARNING opportunity for memdup_user

vim +8516 fs/io_uring.c

  8497	
  8498	static int io_register_restrictions(struct io_ring_ctx *ctx, void __user *arg,
  8499					    unsigned int nr_args)
  8500	{
  8501		struct io_uring_restriction *res;
  8502		size_t size;
  8503		int i, ret;
  8504	
  8505		/* We allow only a single restrictions registration */
  8506		if (ctx->restricted)
  8507			return -EBUSY;
  8508	
  8509		if (!arg || nr_args > IORING_MAX_RESTRICTIONS)
  8510			return -EINVAL;
  8511	
  8512		size = array_size(nr_args, sizeof(*res));
  8513		if (size == SIZE_MAX)
  8514			return -EOVERFLOW;
  8515	
> 8516		res = kmalloc(size, GFP_KERNEL);
  8517		if (!res)
  8518			return -ENOMEM;
  8519	
  8520		if (copy_from_user(res, arg, size)) {
  8521			ret = -EFAULT;
  8522			goto out;
  8523		}
  8524	
  8525		for (i = 0; i < nr_args; i++) {
  8526			switch (res[i].opcode) {
  8527			case IORING_RESTRICTION_REGISTER_OP:
  8528				if (res[i].register_op >= IORING_REGISTER_LAST) {
  8529					ret = -EINVAL;
  8530					goto out;
  8531				}
  8532	
  8533				__set_bit(res[i].register_op,
  8534					  ctx->restrictions.register_op);
  8535				break;
  8536			case IORING_RESTRICTION_SQE_OP:
  8537				if (res[i].sqe_op >= IORING_OP_LAST) {
  8538					ret = -EINVAL;
  8539					goto out;
  8540				}
  8541	
  8542				__set_bit(res[i].sqe_op, ctx->restrictions.sqe_op);
  8543				break;
  8544			case IORING_RESTRICTION_SQE_FLAGS_ALLOWED:
  8545				ctx->restrictions.sqe_flags_allowed = res[i].sqe_flags;
  8546				break;
  8547			case IORING_RESTRICTION_SQE_FLAGS_REQUIRED:
  8548				ctx->restrictions.sqe_flags_required = res[i].sqe_flags;
  8549				break;
  8550			default:
  8551				ret = -EINVAL;
  8552				goto out;
  8553			}
  8554		}
  8555	
  8556		ctx->restricted = 1;
  8557	
  8558		ret = 0;
  8559	out:
  8560		/* Reset all restrictions if an error happened */
  8561		if (ret != 0)
  8562			memset(&ctx->restrictions, 0, sizeof(ctx->restrictions));
  8563	
  8564		kfree(res);
  8565		return ret;
  8566	}
  8567	

---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
Stefano Garzarella Aug. 17, 2020, 10:41 a.m. UTC | #2 
On Fri, Aug 14, 2020 at 01:42:15AM +0800, kernel test robot wrote:
> Hi Stefano,
> 
> I love your patch! Perhaps something to improve:
> 
> [auto build test WARNING on linus/master]
> [also build test WARNING on v5.8 next-20200813]
> [If your patch is applied to the wrong git tree, kindly drop us a note.
> And when submitting patch, we suggest to use '--base' as documented in
> https://git-scm.com/docs/git-format-patch]
> 
> url:    https://github.com/0day-ci/linux/commits/Stefano-Garzarella/io_uring-add-restrictions-to-support-untrusted-applications-and-guests/20200813-233653
> base:   https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git dc06fe51d26efc100ac74121607c01a454867c91
> config: s390-randconfig-c003-20200813 (attached as .config)
> compiler: s390-linux-gcc (GCC) 9.3.0
> 
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <lkp@intel.com>
> 
> 
> coccinelle warnings: (new ones prefixed by >>)
> 
> >> fs/io_uring.c:8516:7-14: WARNING opportunity for memdup_user

Yeah, I think make sense.

I'll use memdup_user() in the next version.

> 
> vim +8516 fs/io_uring.c
> 
>   8497	
>   8498	static int io_register_restrictions(struct io_ring_ctx *ctx, void __user *arg,
>   8499					    unsigned int nr_args)
>   8500	{
>   8501		struct io_uring_restriction *res;
>   8502		size_t size;
>   8503		int i, ret;
>   8504	
>   8505		/* We allow only a single restrictions registration */
>   8506		if (ctx->restricted)
>   8507			return -EBUSY;
>   8508	
>   8509		if (!arg || nr_args > IORING_MAX_RESTRICTIONS)
>   8510			return -EINVAL;
>   8511	
>   8512		size = array_size(nr_args, sizeof(*res));
>   8513		if (size == SIZE_MAX)
>   8514			return -EOVERFLOW;
>   8515	
> > 8516		res = kmalloc(size, GFP_KERNEL);
>   8517		if (!res)
>   8518			return -ENOMEM;
>   8519	
>   8520		if (copy_from_user(res, arg, size)) {
>   8521			ret = -EFAULT;
>   8522			goto out;
>   8523		}
>   8524	
>   8525		for (i = 0; i < nr_args; i++) {
>   8526			switch (res[i].opcode) {
>   8527			case IORING_RESTRICTION_REGISTER_OP:
>   8528				if (res[i].register_op >= IORING_REGISTER_LAST) {
>   8529					ret = -EINVAL;
>   8530					goto out;
>   8531				}
>   8532	
>   8533				__set_bit(res[i].register_op,
>   8534					  ctx->restrictions.register_op);
>   8535				break;
>   8536			case IORING_RESTRICTION_SQE_OP:
>   8537				if (res[i].sqe_op >= IORING_OP_LAST) {
>   8538					ret = -EINVAL;
>   8539					goto out;
>   8540				}
>   8541	
>   8542				__set_bit(res[i].sqe_op, ctx->restrictions.sqe_op);
>   8543				break;
>   8544			case IORING_RESTRICTION_SQE_FLAGS_ALLOWED:
>   8545				ctx->restrictions.sqe_flags_allowed = res[i].sqe_flags;
>   8546				break;
>   8547			case IORING_RESTRICTION_SQE_FLAGS_REQUIRED:
>   8548				ctx->restrictions.sqe_flags_required = res[i].sqe_flags;
>   8549				break;
>   8550			default:
>   8551				ret = -EINVAL;
>   8552				goto out;
>   8553			}
>   8554		}
>   8555	
>   8556		ctx->restricted = 1;
>   8557	
>   8558		ret = 0;
>   8559	out:
>   8560		/* Reset all restrictions if an error happened */
>   8561		if (ret != 0)
>   8562			memset(&ctx->restrictions, 0, sizeof(ctx->restrictions));
>   8563	
>   8564		kfree(res);
>   8565		return ret;
>   8566	}
>   8567	
> 
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
Kees Cook Aug. 26, 2020, 7:46 p.m. UTC | #3 
On Thu, Aug 13, 2020 at 05:32:53PM +0200, Stefano Garzarella wrote:
> +/*
> + * io_uring_restriction->opcode values
> + */
> +enum {
> +	/* Allow an io_uring_register(2) opcode */
> +	IORING_RESTRICTION_REGISTER_OP,
> +
> +	/* Allow an sqe opcode */
> +	IORING_RESTRICTION_SQE_OP,
> +
> +	/* Allow sqe flags */
> +	IORING_RESTRICTION_SQE_FLAGS_ALLOWED,
> +
> +	/* Require sqe flags (these flags must be set on each submission) */
> +	IORING_RESTRICTION_SQE_FLAGS_REQUIRED,
> +
> +	IORING_RESTRICTION_LAST
> +};

Same thought on enum literals, but otherwise, looks good:

Reviewed-by: Kees Cook <keescook@chromium.org>
Stefano Garzarella Aug. 27, 2020, 7:12 a.m. UTC | #4 
On Wed, Aug 26, 2020 at 12:46:24PM -0700, Kees Cook wrote:
> On Thu, Aug 13, 2020 at 05:32:53PM +0200, Stefano Garzarella wrote:
> > +/*
> > + * io_uring_restriction->opcode values
> > + */
> > +enum {
> > +	/* Allow an io_uring_register(2) opcode */
> > +	IORING_RESTRICTION_REGISTER_OP,
> > +
> > +	/* Allow an sqe opcode */
> > +	IORING_RESTRICTION_SQE_OP,
> > +
> > +	/* Allow sqe flags */
> > +	IORING_RESTRICTION_SQE_FLAGS_ALLOWED,
> > +
> > +	/* Require sqe flags (these flags must be set on each submission) */
> > +	IORING_RESTRICTION_SQE_FLAGS_REQUIRED,
> > +
> > +	IORING_RESTRICTION_LAST
> > +};
> 
> Same thought on enum literals, but otherwise, looks good:

Sure, I'll fix the enum in the next version.

> 
> Reviewed-by: Kees Cook <keescook@chromium.org>

Thanks for the review,
Stefano
diff mbox series

Patch

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 1ec25ee71372..cb365e6e0af7 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -98,6 +98,8 @@ 
 #define IORING_MAX_FILES_TABLE	(1U << IORING_FILE_TABLE_SHIFT)
 #define IORING_FILE_TABLE_MASK	(IORING_MAX_FILES_TABLE - 1)
 #define IORING_MAX_FIXED_FILES	(64 * IORING_MAX_FILES_TABLE)
+#define IORING_MAX_RESTRICTIONS	(IORING_RESTRICTION_LAST + \
+				 IORING_REGISTER_LAST + IORING_OP_LAST)
 
 struct io_uring {
 	u32 head ____cacheline_aligned_in_smp;
@@ -219,6 +221,13 @@  struct io_buffer {
 	__u16 bid;
 };
 
+struct io_restriction {
+	DECLARE_BITMAP(register_op, IORING_REGISTER_LAST);
+	DECLARE_BITMAP(sqe_op, IORING_OP_LAST);
+	u8 sqe_flags_allowed;
+	u8 sqe_flags_required;
+};
+
 struct io_ring_ctx {
 	struct {
 		struct percpu_ref	refs;
@@ -231,6 +240,7 @@  struct io_ring_ctx {
 		unsigned int		cq_overflow_flushed: 1;
 		unsigned int		drain_next: 1;
 		unsigned int		eventfd_async: 1;
+		unsigned int		restricted: 1;
 
 		/*
 		 * Ring buffer of indices into array of io_uring_sqe, which is
@@ -338,6 +348,7 @@  struct io_ring_ctx {
 	struct llist_head		file_put_llist;
 
 	struct work_struct		exit_work;
+	struct io_restriction		restrictions;
 };
 
 /*
@@ -6353,6 +6364,19 @@  static int io_init_req(struct io_ring_ctx *ctx, struct io_kiocb *req,
 	if (unlikely(sqe_flags & ~SQE_VALID_FLAGS))
 		return -EINVAL;
 
+	if (unlikely(ctx->restricted)) {
+		if (!test_bit(req->opcode, ctx->restrictions.sqe_op))
+			return -EACCES;
+
+		if ((sqe_flags & ctx->restrictions.sqe_flags_required) !=
+		    ctx->restrictions.sqe_flags_required)
+			return -EACCES;
+
+		if (sqe_flags & ~(ctx->restrictions.sqe_flags_allowed |
+				  ctx->restrictions.sqe_flags_required))
+			return -EACCES;
+	}
+
 	if ((sqe_flags & IOSQE_BUFFER_SELECT) &&
 	    !io_op_defs[req->opcode].buffer_select)
 		return -EOPNOTSUPP;
@@ -8650,6 +8674,76 @@  static int io_unregister_personality(struct io_ring_ctx *ctx, unsigned id)
 	return -EINVAL;
 }
 
+static int io_register_restrictions(struct io_ring_ctx *ctx, void __user *arg,
+				    unsigned int nr_args)
+{
+	struct io_uring_restriction *res;
+	size_t size;
+	int i, ret;
+
+	/* We allow only a single restrictions registration */
+	if (ctx->restricted)
+		return -EBUSY;
+
+	if (!arg || nr_args > IORING_MAX_RESTRICTIONS)
+		return -EINVAL;
+
+	size = array_size(nr_args, sizeof(*res));
+	if (size == SIZE_MAX)
+		return -EOVERFLOW;
+
+	res = kmalloc(size, GFP_KERNEL);
+	if (!res)
+		return -ENOMEM;
+
+	if (copy_from_user(res, arg, size)) {
+		ret = -EFAULT;
+		goto out;
+	}
+
+	for (i = 0; i < nr_args; i++) {
+		switch (res[i].opcode) {
+		case IORING_RESTRICTION_REGISTER_OP:
+			if (res[i].register_op >= IORING_REGISTER_LAST) {
+				ret = -EINVAL;
+				goto out;
+			}
+
+			__set_bit(res[i].register_op,
+				  ctx->restrictions.register_op);
+			break;
+		case IORING_RESTRICTION_SQE_OP:
+			if (res[i].sqe_op >= IORING_OP_LAST) {
+				ret = -EINVAL;
+				goto out;
+			}
+
+			__set_bit(res[i].sqe_op, ctx->restrictions.sqe_op);
+			break;
+		case IORING_RESTRICTION_SQE_FLAGS_ALLOWED:
+			ctx->restrictions.sqe_flags_allowed = res[i].sqe_flags;
+			break;
+		case IORING_RESTRICTION_SQE_FLAGS_REQUIRED:
+			ctx->restrictions.sqe_flags_required = res[i].sqe_flags;
+			break;
+		default:
+			ret = -EINVAL;
+			goto out;
+		}
+	}
+
+	ctx->restricted = 1;
+
+	ret = 0;
+out:
+	/* Reset all restrictions if an error happened */
+	if (ret != 0)
+		memset(&ctx->restrictions, 0, sizeof(ctx->restrictions));
+
+	kfree(res);
+	return ret;
+}
+
 static bool io_register_op_must_quiesce(int op)
 {
 	switch (op) {
@@ -8696,6 +8790,18 @@  static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 		if (ret) {
 			percpu_ref_resurrect(&ctx->refs);
 			ret = -EINTR;
+			goto out_quiesce;
+		}
+	}
+
+	if (ctx->restricted) {
+		if (opcode >= IORING_REGISTER_LAST) {
+			ret = -EINVAL;
+			goto out;
+		}
+
+		if (!test_bit(opcode, ctx->restrictions.register_op)) {
+			ret = -EACCES;
 			goto out;
 		}
 	}
@@ -8759,15 +8865,19 @@  static int __io_uring_register(struct io_ring_ctx *ctx, unsigned opcode,
 			break;
 		ret = io_unregister_personality(ctx, nr_args);
 		break;
+	case IORING_REGISTER_RESTRICTIONS:
+		ret = io_register_restrictions(ctx, arg, nr_args);
+		break;
 	default:
 		ret = -EINVAL;
 		break;
 	}
 
+out:
 	if (io_register_op_must_quiesce(opcode)) {
 		/* bring the ctx back to life */
 		percpu_ref_reinit(&ctx->refs);
-out:
+out_quiesce:
 		reinit_completion(&ctx->ref_comp);
 	}
 	return ret;
diff --git a/include/uapi/linux/io_uring.h b/include/uapi/linux/io_uring.h
index cdc98afbacc3..be54bc3cf173 100644
--- a/include/uapi/linux/io_uring.h
+++ b/include/uapi/linux/io_uring.h
@@ -267,6 +267,7 @@  enum {
 	IORING_REGISTER_PROBE,
 	IORING_REGISTER_PERSONALITY,
 	IORING_UNREGISTER_PERSONALITY,
+	IORING_REGISTER_RESTRICTIONS,
 
 	/* this goes last */
 	IORING_REGISTER_LAST
@@ -295,4 +296,34 @@  struct io_uring_probe {
 	struct io_uring_probe_op ops[0];
 };
 
+struct io_uring_restriction {
+	__u16 opcode;
+	union {
+		__u8 register_op; /* IORING_RESTRICTION_REGISTER_OP */
+		__u8 sqe_op;      /* IORING_RESTRICTION_SQE_OP */
+		__u8 sqe_flags;   /* IORING_RESTRICTION_SQE_FLAGS_* */
+	};
+	__u8 resv;
+	__u32 resv2[3];
+};
+
+/*
+ * io_uring_restriction->opcode values
+ */
+enum {
+	/* Allow an io_uring_register(2) opcode */
+	IORING_RESTRICTION_REGISTER_OP,
+
+	/* Allow an sqe opcode */
+	IORING_RESTRICTION_SQE_OP,
+
+	/* Allow sqe flags */
+	IORING_RESTRICTION_SQE_FLAGS_ALLOWED,
+
+	/* Require sqe flags (these flags must be set on each submission) */
+	IORING_RESTRICTION_SQE_FLAGS_REQUIRED,
+
+	IORING_RESTRICTION_LAST
+};
+
 #endif