Re: [PATCH v2 13/15] PCI: rcar-gen4: Route Root Port AER to a virtual Root Port IRQ

[Marek Vasut]

On 9/28/26 6:52 PM, Koichiro Den wrote:
> On R-Car Gen4, the APP block exposes the Root Port interrupt signals in
> PCIEINTSTS0 and routes them to platform IRQs: AER to intreq_pcim_lperr,
> and PME and bandwidth notifications to intreq_pcim_sub. On an R-Car S4
> with GIC ITS, a link retrain set LnkSta.LBMS and latched PCIEINTSTS0[23]
> (cfg_bw_mgt_msi), but the Root Port's MSI never fired while endpoint
> MSIs worked fine. With iMSI-RX the Root Port MSI capabilities are
> already hidden, but its INTx maps to intreq_pcim_sub. Note that even
> before this driver took over the IRQ with
> rcar_gen4_pcie_msi_irq_handler, the port service was unable to request
> it, as the DWC core used that IRQ for its chained MSI handler. In either
> case (ie. iMSI-RX or an external MSI controller), Root Port interrupts
> are not delivered to the port services.
>
> Set no_rp_msi so the port services always fall back to INTx. Allocate an
> IRQ in a single-entry domain and return its number from the bridge's
> map_irq callback for the Root Port only, leaving downstream INTx
> mappings unchanged. Notifications from different platform IRQs can then
> feed the one IRQ the port services share. Route the AER notifications
> from the "aer" interrupt to it. rcar_gen4_pcie_host_msi_init() sets
> MSICAP0.MSIE when pci_msi_enabled() is true, and hiding the capability
> does not clear it, so handle both the MSI and INTx forms of the status
> bits. DTs without "aer" remain supported, but the APP AER notification
> sources stay masked.
>
> Preserve the virtual IRQ's enable state across controller resets, but
> keep its APP sources masked while reinitialization is pending or Root
> Port events are disarmed. Clear only the MSI-form APP latch, before the
> AER handler clears Root Error Status, to avoid clearing a newly latched
> notification.
>
> Handle pending reset requests before dispatching AER to avoid unsafe DBI
> access. Serialize dispatches from the parent IRQs, including the "msi"
> path added by the next patch, since handle_level_irq() could otherwise
> acknowledge a concurrent interrupt without running its handlers. Recheck
> reset requests after taking the lock in case the other parent has marked
> reinitialization as pending.
>
> Signed-off-by: Koichiro Den <den@xxxxxxxxxxxxx>

Reviewed-by: Marek Vasut <marek.vasut+renesas@xxxxxxxxxxx>