Hi Bjorn,
On Wed, 2023-12-06 at 11:07 -0600, Bjorn Helgaas wrote:
> On Wed, Dec 06, 2023 at 04:37:52PM +0800, Jianjun Wang wrote:
> > Use 'dmam_alloc_coherent' to allocate the MSI address, instead of
> using
> > 'virt_to_phys'.
>
> s/'dmam_alloc_coherent'/dmam_alloc_coherent()/
> s/'virt_to_phys'/virt_to_phys()/
>
> In subject also.
>
> > @@ -732,8 +740,11 @@ static int mtk_pcie_startup_port_v2(struct
> mtk_pcie_port *port)
> > val &= ~INTX_MASK;
> > writel(val, port->base + PCIE_INT_MASK);
> >
> > -if (IS_ENABLED(CONFIG_PCI_MSI))
> > -mtk_pcie_enable_msi(port);
> > +if (IS_ENABLED(CONFIG_PCI_MSI)) {
> > +err = mtk_pcie_enable_msi(port);
> > +if (err)
> > +return err;
>
> Is failure to enable MSI a fatal issue? It looks like this will make
> the host controller completely unusable if we can't set up MSI, even
> if downstream PCI devices could use INTx and get along without MSI.
This shouldn't be a fatal issue, we can still use INTx, I'll fix them
in the next version, thanks for your review.
Thanks.
>
> Bjorn
