[PATCH for-next] io_uring: return EBADFD when ring isn't in the right state

From: Stefano Garzarella <sgarzare@redhat.com>
To: Jens Axboe <axboe@kernel.dk>
Cc: io-uring@vger.kernel.org,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org
Subject: [PATCH for-next] io_uring: return EBADFD when ring isn't in the right state
Date: Tue,  8 Sep 2020 18:52:42 +0200	[thread overview]
Message-ID: <20200908165242.124957-1-sgarzare@redhat.com> (raw)

This patch uniforms the returned error (EBADFD) when the ring state
(enabled/disabled) is not the expected one.

The changes affect io_uring_enter() and io_uring_register() syscalls.

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
---

Hi Jens,
I also updated the test/register-restrictions in liburing here:

https://github.com/stefano-garzarella/liburing (branch: fix-disabled-ring-error)

Thanks,
Stefano
---
 fs/io_uring.c | 5 +++--
 1 file changed, 3 insertions(+), 2 deletions(-)

diff --git a/fs/io_uring.c b/fs/io_uring.c
index 0490edfcdd88..cf5992e79d88 100644
--- a/fs/io_uring.c
+++ b/fs/io_uring.c
@@ -8642,6 +8642,7 @@ SYSCALL_DEFINE6(io_uring_enter, unsigned int, fd, u32, to_submit,
 	if (!percpu_ref_tryget(&ctx->refs))
 		goto out_fput;
 
+	ret = -EBADFD;
 	if (ctx->flags & IORING_SETUP_R_DISABLED)
 		goto out_fput;
 
@@ -9137,7 +9138,7 @@ static int io_register_restrictions(struct io_ring_ctx *ctx, void __user *arg,
 
 	/* Restrictions allowed only if rings started disabled */
 	if (!(ctx->flags & IORING_SETUP_R_DISABLED))
-		return -EINVAL;
+		return -EBADFD;
 
 	/* We allow only a single restrictions registration */
 	if (ctx->restrictions.registered)
@@ -9201,7 +9202,7 @@ static int io_register_restrictions(struct io_ring_ctx *ctx, void __user *arg,
 static int io_register_enable_rings(struct io_ring_ctx *ctx)
 {
 	if (!(ctx->flags & IORING_SETUP_R_DISABLED))
-		return -EINVAL;
+		return -EBADFD;
 
 	if (ctx->restrictions.registered)
 		ctx->restricted = 1;
-- 
2.26.2

