| Message ID | 20250324200540.910962-2-csander@purestorage.com (mailing list archive) |
|---|---|
| State | New |
| Headers | show |
| Series | nvme_map_user_request() cleanup | expand |
On 3/24/25 2:05 PM, Caleb Sander Mateos wrote: > The vectorized io_uring NVMe passthru opcodes don't yet support fixed > buffers. But since userspace can trigger this condition based on the > io_uring SQE parameters, it shouldn't cause a kernel warning. Looks good to me: Reviewed-by: Jens Axboe <axboe@kernel.dk>
diff --git a/drivers/nvme/host/ioctl.c b/drivers/nvme/host/ioctl.c index a35ff018da74..0634e24eac97 100644 --- a/drivers/nvme/host/ioctl.c +++ b/drivers/nvme/host/ioctl.c @@ -140,11 +140,11 @@ static int nvme_map_user_request(struct request *req, u64 ubuffer, if (ioucmd && (ioucmd->flags & IORING_URING_CMD_FIXED)) { struct iov_iter iter; /* fixedbufs is only for non-vectored io */ - if (WARN_ON_ONCE(flags & NVME_IOCTL_VEC)) { + if (flags & NVME_IOCTL_VEC) { ret = -EINVAL; goto out; } ret = io_uring_cmd_import_fixed(ubuffer, bufflen, rq_data_dir(req), &iter, ioucmd,
The vectorized io_uring NVMe passthru opcodes don't yet support fixed buffers. But since userspace can trigger this condition based on the io_uring SQE parameters, it shouldn't cause a kernel warning. Signed-off-by: Caleb Sander Mateos <csander@purestorage.com> Fixes: 23fd22e55b76 ("nvme: wire up fixed buffer support for nvme passthrough") --- drivers/nvme/host/ioctl.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)