[PATCH v2] PCI: rockchip: Avoid accessing PCIe registers with clocks gated

[Bjorn Helgaas]

On Tue, Jun 29, 2021 at 11:52:44AM +0100, Robin Murphy wrote:

> Well, it does use devm_request_irq() so the handler should be unregistered
> by devres *after* ->remove has finished, however that does still leave a
> potential race window in which a pending IRQ could be taken during the later
> part of rockchip_pcie_remove() after it has started turning off critical
> things. Unless the clocks and regulators can also be delegated to devres, it
> might be more robust to explicitly manage the IRQs as well. Mixing the two
> schemes can be problematic when the exact order of both setup and teardown
> matters.

Thanks for this.  I missed this problem.  We have lots of PCI
controller drivers that use some devm interfaces but use the non-devm
clk_prepare_enable(), and they generally turn things off in their
.remove() methods, which is obviously before any devm unregistration.

Many .suspend() methods turn off clocks and power while leaving the
IRQ handler installed.  Should we get an interrupt from our device
after .suspend()?  *Probably* not, but it makes me a little queasy to
rely on that, or to rely on the assumption that the IRQ is not shared.

Bjorn