[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <09f8a530-1bd0-4dc3-85a2-8d67fdcd0148@starfivetech.com>
Date: Thu, 26 Oct 2023 10:22:34 +0800
From: Minda Chen <minda.chen@...rfivetech.com>
To: Conor Dooley <conor@...nel.org>
CC: Krzysztof WilczyĆski <kw@...ux.com>,
Rob Herring <robh+dt@...nel.org>,
Bjorn Helgaas <bhelgaas@...gle.com>,
Lorenzo Pieralisi <lpieralisi@...nel.org>,
Daire McNamara <daire.mcnamara@...rochip.com>,
"Emil Renner Berthing" <emil.renner.berthing@...onical.com>,
Krzysztof Kozlowski <krzysztof.kozlowski+dt@...aro.org>,
<devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<linux-riscv@...ts.infradead.org>, <linux-pci@...r.kernel.org>,
Paul Walmsley <paul.walmsley@...ive.com>,
"Palmer Dabbelt" <palmer@...belt.com>,
Albert Ou <aou@...s.berkeley.edu>,
"Philipp Zabel" <p.zabel@...gutronix.de>,
Mason Huo <mason.huo@...rfivetech.com>,
Leyfoon Tan <leyfoon.tan@...rfivetech.com>,
Kevin Xie <kevin.xie@...rfivetech.com>
Subject: Re: [PATCH v9 14/20] PCI: microchip: Add get_events() callback
function
On 2023/10/25 18:44, Conor Dooley wrote:
> Hey,
>
> On Fri, Oct 20, 2023 at 06:43:35PM +0800, Minda Chen wrote:
>> PolarFire implements their own PCIe interrupts,
>> which added to global PCIe field for PLDA lack of
>> MSI controller, the interrupts to event num mapping
>> is different to PLDA local interrupts. So add
>> get_events() function pointer.
>
> Just FYI, it's not the PLDA IP's lack of an MSI controller, it's the SoC
> itself that doesn't have one. Last time I spoke to Daire about this, he
> was surprised that you didn't need something similar. I might reword
> this as
>
> "PolarFire SoC implements its own PCIe interrupts, additional to the
> regular PCIe interrupts, due to the lack of an MSI controller, so the
> interrupt to event number mapping is different to the PLDA local
> interrupts, necessitating a custom get_events() implementation."
>
>> Also add struct plda_event_ops function pointer structure
>> to struct plda_pcie_rp.
>
> I'd probably also drop this, as it is evident from the diff.
>
> Cheers,
> Conor.
>
OK. Thanks very much.
>> plda_handle_events() will call the get_events() callback
>> function pointer directly. For the robustness of codes,
>> add checking in plda_init_interrupts().
>>
>> Signed-off-by: Minda Chen <minda.chen@...rfivetech.com>
>> Acked-by: Conor Dooley <conor.dooley@...rochip.com>
>> ---
>> drivers/pci/controller/plda/pcie-microchip-host.c | 14 +++++++++++++-
>> drivers/pci/controller/plda/pcie-plda.h | 8 ++++++++
>> 2 files changed, 21 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/pci/controller/plda/pcie-microchip-host.c b/drivers/pci/controller/plda/pcie-microchip-host.c
>> index e57827bdb4b3..5a8c134bf643 100644
>> --- a/drivers/pci/controller/plda/pcie-microchip-host.c
>> +++ b/drivers/pci/controller/plda/pcie-microchip-host.c
>> @@ -652,7 +652,7 @@ static void plda_handle_event(struct irq_desc *desc)
>>
>> chained_irq_enter(chip, desc);
>>
>> - events = mc_get_events(port);
>> + events = port->event_ops->get_events(port);
>>
>> for_each_set_bit(bit, &events, port->num_events)
>> generic_handle_domain_irq(port->event_domain, bit);
>> @@ -811,7 +811,12 @@ static int mc_request_event_irq(struct plda_pcie_rp *plda, int event_irq,
>> 0, event_cause[event].sym, plda);
>> }
>>
>> +static const struct plda_event_ops mc_event_ops = {
>> + .get_events = mc_get_events,
>> +};
>> +
>> static const struct plda_event mc_event = {
>> + .event_ops = &mc_event_ops,
>> .request_event_irq = mc_request_event_irq,
>> .intx_event = EVENT_LOCAL_PM_MSI_INT_INTX,
>> .msi_event = EVENT_LOCAL_PM_MSI_INT_MSI,
>> @@ -925,6 +930,11 @@ static int plda_init_interrupts(struct platform_device *pdev,
>> int i, intx_irq, msi_irq, event_irq;
>> int ret;
>>
>> + if (!event->event_ops || !event->event_ops->get_events) {
>> + dev_err(dev, "no get events ops\n");
>> + return -EINVAL;
>> + }
>> +
>> ret = plda_pcie_init_irq_domains(port);
>> if (ret) {
>> dev_err(dev, "failed creating IRQ domains\n");
>> @@ -935,6 +945,8 @@ static int plda_init_interrupts(struct platform_device *pdev,
>> if (irq < 0)
>> return -ENODEV;
>>
>> + port->event_ops = event->event_ops;
>> +
>> for (i = 0; i < port->num_events; i++) {
>> event_irq = irq_create_mapping(port->event_domain, i);
>> if (!event_irq) {
>> diff --git a/drivers/pci/controller/plda/pcie-plda.h b/drivers/pci/controller/plda/pcie-plda.h
>> index fba7343f9a96..df1729095952 100644
>> --- a/drivers/pci/controller/plda/pcie-plda.h
>> +++ b/drivers/pci/controller/plda/pcie-plda.h
>> @@ -102,6 +102,12 @@
>> #define EVENT_PM_MSI_INT_SYS_ERR 12
>> #define NUM_PLDA_EVENTS 13
>>
>> +struct plda_pcie_rp;
>> +
>> +struct plda_event_ops {
>> + u32 (*get_events)(struct plda_pcie_rp *pcie);
>> +};
>> +
>> struct plda_msi {
>> struct mutex lock; /* Protect used bitmap */
>> struct irq_domain *msi_domain;
>> @@ -117,11 +123,13 @@ struct plda_pcie_rp {
>> struct irq_domain *event_domain;
>> raw_spinlock_t lock;
>> struct plda_msi msi;
>> + const struct plda_event_ops *event_ops;
>> void __iomem *bridge_addr;
>> int num_events;
>> };
>>
>> struct plda_event {
>> + const struct plda_event_ops *event_ops;
>> int (*request_event_irq)(struct plda_pcie_rp *pcie,
>> int event_irq, int event);
>> int intx_event;
>> --
>> 2.17.1
>>
Powered by blists - more mailing lists