Re: [PATCH v3 4/8] iommufd: Add iommufd fault object

From: Baolu Lu
Date: Mon Mar 25 2024 - 08:50:11 EST


On 3/25/24 11:26 AM, Baolu Lu wrote:
On 3/23/24 1:22 AM, Jason Gunthorpe wrote:
On Wed, Mar 20, 2024 at 04:18:05PM +0000, Shameerali Kolothum Thodi wrote:
What I have noticed is that,
-read interface works fine and I can receive struct tiommu_hwpt_pgfault data.
-But once Guest handles the page faults and returns the page response,
  the write to fault fd never reaches the kernel. The sequence is like below,
   sqe = io_uring_get_sqe(ring);
   io_uring_prep_write(sqe, hwpt->fault_fd, resp, sizeof(*resp), 0);
   io_uring_sqe_set_data(sqe, resp);
   io_uring_submit(ring);
   ret = io_uring_wait_cqe(ring, &cqe);
   ....
Please find the function here[2]

The above cqe wait never returns and hardware times out without receiving
page response. My understanding of io_uring default op is that it tries to
issue an sqe as non-blocking first. But it looks like the above write sequence
ends up in kernel poll_wait() as well.Not sure how we can avoid that for
write.
Ah, right, it is because poll can't be choosy about read/write, it has
to work equally for both directions. iommufd_fault_fops_poll() never
returns EPOLLOUT

It should just always return EPOLLOUT because we don't have any queue
to manage.

Are you suggesting the poll file operation to be like below?

static __poll_t iommufd_fault_fops_poll(struct file *filep,
                                        struct poll_table_struct *wait)
{
        struct iommufd_fault *fault = filep->private_data;
        __poll_t pollflags = EPOLLOUT;

        poll_wait(filep, &fault->wait_queue, wait);
        mutex_lock(&fault->mutex);
        if (!list_empty(&fault->deliver))
                pollflags = EPOLLIN | EPOLLRDNORM;

If I understood it correctly, here it should be,

pollflags |= EPOLLIN | EPOLLRDNORM;

        mutex_unlock(&fault->mutex);

        return pollflags;
}

Best regards,
baolu