Re: [PATCH V3 1/2] PCI/DOE: Remove the pci_doe_flush_mb() call

From: Ira Weiny
Date: Mon Nov 28 2022 - 12:53:38 EST


On Mon, Nov 28, 2022 at 01:51:24PM +0800, Li, Ming wrote:
>
> On 11/28/2022 12:03 PM, ira.weiny@xxxxxxxxx wrote:
> > From: Ira Weiny <ira.weiny@xxxxxxxxx>
> >
> > pci_doe_flush_mb() does not work and is currently unused.
> >
> > It does not work because each struct doe_mb is managed as part of the
> > PCI device. They can't go away as long as the PCI device exists.
> > pci_doe_flush_mb() was set up to flush the workqueue and prevent any
> > further submissions to the mailboxes when the PCI device goes away.
> > Unfortunately, this was fundamentally flawed. There was no guarantee
> > that a struct doe_mb remained after pci_doe_flush_mb() returned.
> > Therefore, the doe_mb state could be invalid when those threads waiting
> > on the workqueue were flushed.
> >
> > Fortunately the current code is safe because all callers make a
> > synchronous call to pci_doe_submit_task() and maintain a reference on
> > the PCI device. Therefore pci_doe_flush_mb() is effectively unused.
> >
> > Rather than attempt to fix pci_doe_flush_mb() just remove the dead code
> > around pci_doe_flush_mb().
> >
> > Cc: Lukas Wunner <lukas@xxxxxxxxx>
> > Cc: Jonathan Cameron <Jonathan.Cameron@xxxxxxxxxx>
> > Signed-off-by: Ira Weiny <ira.weiny@xxxxxxxxx>
>
> Some comments inline.
>
> >
> > ---
> > Changes from V2:
> > Lukas
> > Clarify commit message.
> > Jonathan
> > Add comment for changed poll interval.
>
> ...
>
> >
> > -static int pci_doe_wait(struct pci_doe_mb *doe_mb, unsigned long timeout)
> > -{
> > - if (wait_event_timeout(doe_mb->wq,
> > - test_bit(PCI_DOE_FLAG_CANCEL, &doe_mb->flags),
> > - timeout))
> > - return -EIO;
> > - return 0;
> > -}
> > -
> > static void pci_doe_write_ctrl(struct pci_doe_mb *doe_mb, u32 val)
> > {
> > struct pci_dev *pdev = doe_mb->pdev;
> > @@ -82,12 +73,9 @@ static int pci_doe_abort(struct pci_doe_mb *doe_mb)
> > pci_doe_write_ctrl(doe_mb, PCI_DOE_CTRL_ABORT);
> >
> > do {
> > - int rc;
> > u32 val;
> >
> > - rc = pci_doe_wait(doe_mb, PCI_DOE_POLL_INTERVAL);
> > - if (rc)
> > - return rc;
> > + msleep_interruptible(PCI_DOE_POLL_INTERVAL_MSECS);
> > pci_read_config_dword(pdev, offset + PCI_DOE_STATUS, &val);
>
> Looks like we don't have to use msleep_interruptible() here, can use msleep() directly?

I don't know. I think your suggestion below holds here too. Unfortunately
that could mask the signal received in the case below. But I think that is
going to be rare enough we could ignore it.

>
> >
> > /* Abort success! */
> > @@ -278,11 +266,7 @@ static void doe_statemachine_work(struct work_struct *work)
> > signal_task_abort(task, -EIO);
> > return;
> > }
> > - rc = pci_doe_wait(doe_mb, PCI_DOE_POLL_INTERVAL);
> > - if (rc) {
> > - signal_task_abort(task, rc);
> > - return;
> > - }
> > + msleep_interruptible(PCI_DOE_POLL_INTERVAL_MSECS);
> > goto retry_resp;
> > }
>
> I guess that you use msleep_interruptible() here for aborting current task when signals come.
> So there should be signal_task_abort() and return when msleep_interruptible() receives a signal.

Yes this makes much more sense. Thanks for looking!
Ira

>
> Thanks
> Ming
>
> >
> > @@ -383,21 +367,6 @@ static void pci_doe_destroy_workqueue(void *mb)
> > destroy_workqueue(doe_mb->work_queue);
> > }
> >
> > -static void pci_doe_flush_mb(void *mb)
> > -{
> > - struct pci_doe_mb *doe_mb = mb;
> > -
> > - /* Stop all pending work items from starting */
> > - set_bit(PCI_DOE_FLAG_DEAD, &doe_mb->flags);
> > -
> > - /* Cancel an in progress work item, if necessary */
> > - set_bit(PCI_DOE_FLAG_CANCEL, &doe_mb->flags);
> > - wake_up(&doe_mb->wq);
> > -
> > - /* Flush all work items */
> > - flush_workqueue(doe_mb->work_queue);
> > -}
> > -
> > /**
> > * pcim_doe_create_mb() - Create a DOE mailbox object
> > *
> > @@ -450,14 +419,6 @@ struct pci_doe_mb *pcim_doe_create_mb(struct pci_dev *pdev, u16 cap_offset)
> > return ERR_PTR(rc);
> > }
> >
> > - /*
> > - * The state machine and the mailbox should be in sync now;
> > - * Set up mailbox flush prior to using the mailbox to query protocols.
> > - */
> > - rc = devm_add_action_or_reset(dev, pci_doe_flush_mb, doe_mb);
> > - if (rc)
> > - return ERR_PTR(rc);
> > -
> > rc = pci_doe_cache_protocols(doe_mb);
> > if (rc) {
> > pci_err(pdev, "[%x] failed to cache protocols : %d\n",
>