Linux-ARM-Kernel Archive on lore.kernel.org
 help / color / Atom feed
From: Vidya Sagar <vidyas@nvidia.com>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: mark.rutland@arm.com, devicetree@vger.kernel.org,
	lorenzo.pieralisi@arm.com, mperttunen@nvidia.com,
	mmaddireddy@nvidia.com, linux-pci@vger.kernel.org,
	catalin.marinas@arm.com, will.deacon@arm.com,
	linux-kernel@vger.kernel.org, robh+dt@kernel.org, kishon@ti.com,
	kthota@nvidia.com, Thierry Reding <thierry.reding@gmail.com>,
	gustavo.pimentel@synopsys.com, jingoohan1@gmail.com,
	linux-tegra@vger.kernel.org, jonathanh@nvidia.com,
	linux-arm-kernel@lists.infradead.org, sagar.tv@gmail.com
Subject: Re: [PATCH V7 02/15] PCI: Disable MSI for Tegra194 root port
Date: Wed, 22 May 2019 13:37:58 +0530
Message-ID: <4a9dadc5-7e16-4893-c7c2-acd7f515e039@nvidia.com> (raw)
In-Reply-To: <20190521193616.GE57618@google.com>

On 5/22/2019 1:06 AM, Bjorn Helgaas wrote:
> On Tue, May 21, 2019 at 10:17:26PM +0530, Vidya Sagar wrote:
>> On 5/21/2019 3:57 PM, Thierry Reding wrote:
>>> On Fri, May 17, 2019 at 06:08:33PM +0530, Vidya Sagar wrote:
>>>> Tegra194 rootports don't generate MSI interrupts for PME events and hence
>>>> MSI needs to be disabled for them to avoid root ports service drivers
>>>> registering their respective ISRs with MSI interrupt.
> 
> The service drivers (AER, hotplug, etc) don't know whether the
> interrupt is INTx or MSI; they just register their ISRs with
> pcie_device.irq.
> 
> The point of this patch is that the PCI core doesn't support devices
> that use MSI and INTx at the same time, and since this device can't
> generate MSI for PME, we have to use INTx for *all* its interrupts.
> 
>>>> Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
>>>> ---
>>>> Changes since [v6]:
>>>> * This is a new patch
>>>>
>>>>    drivers/pci/quirks.c | 14 ++++++++++++++
>>>>    1 file changed, 14 insertions(+)
>>>>
>>>> diff --git a/drivers/pci/quirks.c b/drivers/pci/quirks.c
>>>> index 0f16acc323c6..28f9a0380df5 100644
>>>> --- a/drivers/pci/quirks.c
>>>> +++ b/drivers/pci/quirks.c
>>>> @@ -2592,6 +2592,20 @@ DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA,
>>>>    			PCI_DEVICE_ID_NVIDIA_NVENET_15,
>>>>    			nvenet_msi_disable);
>>>> +/*
>>>> + * Tegra194's PCIe root ports don't generate MSI interrupts for PME events
>>>> + * instead legacy interrupts are generated. Hence, to avoid service drivers
>>>> + * registering their respective ISRs for MSIs, need to disable MSI interrupts
>>>> + * for root ports.
>>>> + */
>>>> +static void disable_tegra194_rp_msi(struct pci_dev *dev)
>>>> +{
>>>> +	dev->no_msi = 1;
>>>> +}
>>>> +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, 0x1ad0, disable_tegra194_rp_msi);
>>>> +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, 0x1ad1, disable_tegra194_rp_msi);
>>>> +DECLARE_PCI_FIXUP_EARLY(PCI_VENDOR_ID_NVIDIA, 0x1ad2, disable_tegra194_rp_msi);
>>>> +
>>>
>>> Later functions in this file seem to use a more consistent naming
>>> pattern, according to which the name for this would become:
>>>
>>> 	pci_quirk_nvidia_tegra194_disable_rp_msi
>>>
>>> Might be worth considering making this consistent.
>>>
>>> This could also be moved to the DWC driver to restrict this to where it
>>> is needed. In either case, this seems like a good solution, so:
>>>
>>> Reviewed-by: Thierry Reding <treding@nvidia.com>
>>>
>> Ok. I'll move it to DWC driver along with name change for the quirk API.
> 
> Is there any possibility this hardware would be used in an ACPI
> system?  If so, the quirk should probably stay in drivers/pci/quirks.c
> because the DWC driver would not be present.
Yes. There is a plan to boot kernel through UEFI (using ACPI) on this system.
So, I'll move it to drivers/pci/quirks.c file.

> 
> Either way, please also add some PCIe spec references about this in
> the changelog and a comment in the code about working around this
> issue.  I think the MSI/MSI-X sections that prohibit a device from
> using both INTx and MSI/MSI-X are probably the most pertinent.
Ok. I'll take care of it in the next patch series.

> 
> The reason I want a comment about this is to discourage future
> hardware from following this example because every device that *does*
> follow this example requires a kernel update that would be otherwise
> unnecessary.
Ok. I'll take it up with our hardware engineers to have only MSI/MSI-X interrupts
getting generated for all root port received events.

> 
> Bjorn
> 


_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel

  reply index

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-05-17 12:38 [PATCH V7 00/15] Add Tegra194 PCIe support Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 01/15] PCI: Add #defines for some of PCIe spec r4.0 features Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 02/15] PCI: Disable MSI for Tegra194 root port Vidya Sagar
2019-05-21 10:27   ` Thierry Reding
2019-05-21 16:47     ` Vidya Sagar
2019-05-21 19:34       ` Vidya Sagar
2019-05-21 19:36       ` Bjorn Helgaas
2019-05-22  8:07         ` Vidya Sagar [this message]
2019-05-17 12:38 ` [PATCH V7 03/15] PCI: dwc: Perform dbi regs write lock towards the end Vidya Sagar
2019-05-21 10:29   ` Thierry Reding
2019-05-17 12:38 ` [PATCH V7 04/15] PCI: dwc: Move config space capability search API Vidya Sagar
2019-05-21 10:29   ` Thierry Reding
2019-05-21 21:17   ` Bjorn Helgaas
2019-05-22  8:56     ` Vidya Sagar
2019-05-22 14:02       ` Bjorn Helgaas
2019-05-24 14:46         ` Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 05/15] PCI: dwc: Add ext " Vidya Sagar
2019-05-21 10:36   ` Thierry Reding
2019-05-21 17:14     ` Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 06/15] dt-bindings: PCI: designware: Add binding for CDM register check Vidya Sagar
2019-05-21 10:37   ` Thierry Reding
2019-05-24 20:23   ` Rob Herring
2019-05-17 12:38 ` [PATCH V7 07/15] PCI: dwc: Add support to enable " Vidya Sagar
2019-05-21 10:38   ` Thierry Reding
2019-05-17 12:38 ` [PATCH V7 08/15] dt-bindings: Add PCIe supports-clkreq property Vidya Sagar
2019-05-21 10:39   ` Thierry Reding
2019-05-17 12:38 ` [PATCH V7 09/15] dt-bindings: PCI: tegra: Add device tree support for Tegra194 Vidya Sagar
2019-05-21 10:51   ` Thierry Reding
2019-05-21 18:00     ` Vidya Sagar
2019-05-24 20:26   ` Rob Herring
2019-05-17 12:38 ` [PATCH V7 10/15] dt-bindings: PHY: P2U: Add Tegra194 P2U block Vidya Sagar
2019-05-21 10:52   ` Thierry Reding
2019-05-17 12:38 ` [PATCH V7 11/15] arm64: tegra: Add P2U and PCIe controller nodes to Tegra194 DT Vidya Sagar
2019-05-17 13:03   ` Ard Biesheuvel
2019-05-17 17:38     ` Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 12/15] arm64: tegra: Enable PCIe slots in P2972-0000 board Vidya Sagar
2019-05-21 10:54   ` Thierry Reding
2019-05-21 18:17     ` Vidya Sagar
2019-05-22 13:48       ` Thierry Reding
2019-05-17 12:38 ` [PATCH V7 13/15] phy: tegra: Add PCIe PIPE2UPHY support Vidya Sagar
2019-05-21 11:00   ` Thierry Reding
2019-05-21 19:37     ` Vidya Sagar
2019-05-21 11:00   ` Thierry Reding
2019-05-22  8:59     ` Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 14/15] PCI: tegra: Add Tegra194 PCIe support Vidya Sagar
2019-05-21 11:41   ` Thierry Reding
2019-05-22 12:05     ` Vidya Sagar
2019-05-22 14:14       ` Thierry Reding
2019-05-24 18:07         ` Vidya Sagar
2019-05-17 12:38 ` [PATCH V7 15/15] arm64: Add Tegra194 PCIe driver to defconfig Vidya Sagar

Reply instructions:

You may reply publically 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=4a9dadc5-7e16-4893-c7c2-acd7f515e039@nvidia.com \
    --to=vidyas@nvidia.com \
    --cc=catalin.marinas@arm.com \
    --cc=devicetree@vger.kernel.org \
    --cc=gustavo.pimentel@synopsys.com \
    --cc=helgaas@kernel.org \
    --cc=jingoohan1@gmail.com \
    --cc=jonathanh@nvidia.com \
    --cc=kishon@ti.com \
    --cc=kthota@nvidia.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=linux-tegra@vger.kernel.org \
    --cc=lorenzo.pieralisi@arm.com \
    --cc=mark.rutland@arm.com \
    --cc=mmaddireddy@nvidia.com \
    --cc=mperttunen@nvidia.com \
    --cc=robh+dt@kernel.org \
    --cc=sagar.tv@gmail.com \
    --cc=thierry.reding@gmail.com \
    --cc=will.deacon@arm.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

Linux-ARM-Kernel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-arm-kernel/0 linux-arm-kernel/git/0.git
	git clone --mirror https://lore.kernel.org/linux-arm-kernel/1 linux-arm-kernel/git/1.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-arm-kernel linux-arm-kernel/ https://lore.kernel.org/linux-arm-kernel \
		linux-arm-kernel@lists.infradead.org infradead-linux-arm-kernel@archiver.kernel.org
	public-inbox-index linux-arm-kernel


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.infradead.lists.linux-arm-kernel


AGPL code for this site: git clone https://public-inbox.org/ public-inbox