linux-pci.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: xxm <xxm@rock-chips.com>
To: Robin Murphy <robin.murphy@arm.com>,
	Bjorn Helgaas <bhelgaas@google.com>,
	Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Cc: linux-pci@vger.kernel.org, Shawn Lin <shawn.lin@rock-chips.com>,
	linux-rockchip@lists.infradead.org
Subject: Re: [PATCH 1/3] PCI: dwc: Skip allocating own MSI domain if using external MSI domain
Date: Wed, 20 Jan 2021 09:53:43 +0800	[thread overview]
Message-ID: <c6e2d613-c398-2ed1-a9b9-bdc31d3a5c20@rock-chips.com> (raw)
In-Reply-To: <e17fcf83-18f1-c2c4-3a0c-a68e74138e15@arm.com>

Hi Robin,

在 2021/1/20 3:47, Robin Murphy 写道:
> On 2021-01-18 09:17, Simon Xue wrote:
>> From: Shawn Lin <shawn.lin@rock-chips.com>
>>
>> On some platform, external MSI domain is using instead of the one
>> created by designware driver. For instance, if using GIC-V3-ITS
>> as a MSI domain, we only need set msi-map in the devicetree but
>> never need any bit in the designware driver to handle MSI stuff.
>> So skip allocating its own MSI domain for that case.
>
> How is this different from the existing pp->has_msi_ctrl? AFAICS, 
> dw_pcie_host_init() won't call dw_pcie_allocate_domains() anyway if an 
> external MSI controller is present.

Thanks for your review, I didn't notice pp->has_msi_ctrl already exist, 
will abandon this patch.

Simon Xue.

>
> Robin.
>
>> Signed-off-by: Shawn Lin <shawn.lin@rock-chips.com>
>> Signed-off-by: Simon Xue <xxm@rock-chips.com>
>> ---
>>   drivers/pci/controller/dwc/pcie-designware-host.c | 10 +++++++++-
>>   drivers/pci/controller/dwc/pcie-designware.h      |  1 +
>>   2 files changed, 10 insertions(+), 1 deletion(-)
>>
>> diff --git a/drivers/pci/controller/dwc/pcie-designware-host.c 
>> b/drivers/pci/controller/dwc/pcie-designware-host.c
>> index 8a84c005f32b..d9d93cab970a 100644
>> --- a/drivers/pci/controller/dwc/pcie-designware-host.c
>> +++ b/drivers/pci/controller/dwc/pcie-designware-host.c
>> @@ -235,6 +235,10 @@ int dw_pcie_allocate_domains(struct pcie_port *pp)
>>       struct dw_pcie *pci = to_dw_pcie_from_pp(pp);
>>       struct fwnode_handle *fwnode = 
>> of_node_to_fwnode(pci->dev->of_node);
>>   +    /* Rely on the external MSI domain */
>> +    if (pp->msi_ext)
>> +        return 0;
>> +
>>       pp->irq_domain = irq_domain_create_linear(fwnode, pp->num_vectors,
>>                              &dw_pcie_msi_domain_ops, pp);
>>       if (!pp->irq_domain) {
>> @@ -258,6 +262,9 @@ int dw_pcie_allocate_domains(struct pcie_port *pp)
>>     static void dw_pcie_free_msi(struct pcie_port *pp)
>>   {
>> +    if (pp->msi_ext)
>> +        return;
>> +
>>       if (pp->msi_irq) {
>>           irq_set_chained_handler(pp->msi_irq, NULL);
>>           irq_set_handler_data(pp->msi_irq, NULL);
>> @@ -359,7 +366,8 @@ int dw_pcie_host_init(struct pcie_port *pp)
>>       if (pci->link_gen < 1)
>>           pci->link_gen = of_pci_get_max_link_speed(np);
>>   -    if (pci_msi_enabled()) {
>> +    if (pci_msi_enabled() &&
>> +        !pp->msi_ext) {
>>           pp->has_msi_ctrl = !(pp->ops->msi_host_init ||
>>                        of_property_read_bool(np, "msi-parent") ||
>>                        of_property_read_bool(np, "msi-map"));
>> diff --git a/drivers/pci/controller/dwc/pcie-designware.h 
>> b/drivers/pci/controller/dwc/pcie-designware.h
>> index 0207840756c4..cf3b0664302a 100644
>> --- a/drivers/pci/controller/dwc/pcie-designware.h
>> +++ b/drivers/pci/controller/dwc/pcie-designware.h
>> @@ -195,6 +195,7 @@ struct pcie_port {
>>       u32            irq_mask[MAX_MSI_CTRLS];
>>       struct pci_host_bridge  *bridge;
>>       raw_spinlock_t        lock;
>> +    int            msi_ext;
>>       DECLARE_BITMAP(msi_irq_in_use, MAX_MSI_IRQS);
>>   };
>>
>
>



      reply	other threads:[~2021-01-20  2:10 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-01-18  9:17 [PATCH 1/3] PCI: dwc: Skip allocating own MSI domain if using external MSI domain Simon Xue
2021-01-18  9:17 ` [PATCH 2/3] dt-bindings: rockchip: Add DesignWare based PCIe controller Simon Xue
2021-01-19 13:07   ` Johan Jonker
2021-01-19 13:14     ` Heiko Stübner
2021-01-19 15:11       ` Johan Jonker
2021-01-19 18:40         ` Robin Murphy
2021-01-20 14:16           ` Rob Herring
2021-01-20 14:54             ` Heiko Stübner
2021-01-20 16:20               ` Robin Murphy
2021-01-19 19:58     ` Robin Murphy
2021-01-20 10:06     ` xxm
2021-01-18  9:17 ` [PATCH 3/3] PCI: rockchip: add " Simon Xue
2021-01-19 20:28   ` Robin Murphy
2021-01-20  1:58     ` xxm
2021-01-19 19:47 ` [PATCH 1/3] PCI: dwc: Skip allocating own MSI domain if using external MSI domain Robin Murphy
2021-01-20  1:53   ` xxm [this message]

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=c6e2d613-c398-2ed1-a9b9-bdc31d3a5c20@rock-chips.com \
    --to=xxm@rock-chips.com \
    --cc=bhelgaas@google.com \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-rockchip@lists.infradead.org \
    --cc=lorenzo.pieralisi@arm.com \
    --cc=robin.murphy@arm.com \
    --cc=shawn.lin@rock-chips.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).