linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Gleixner <tglx@linutronix.de>
To: Marc Zyngier <maz@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>,
	Bjorn Helgaas <helgaas@kernel.org>,
	Alex Williamson <alex.williamson@redhat.com>,
	Kevin Tian <kevin.tian@intel.com>,
	Jason Gunthorpe <jgg@nvidia.com>, Megha Dey <megha.dey@intel.com>,
	Ashok Raj <ashok.raj@intel.com>,
	linux-pci@vger.kernel.org,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	Santosh Shilimkar <ssantosh@kernel.org>,
	iommu@lists.linux-foundation.org, dmaengine@vger.kernel.org,
	Stuart Yoder <stuyoder@gmail.com>,
	Laurentiu Tudor <laurentiu.tudor@nxp.com>,
	Nishanth Menon <nm@ti.com>, Tero Kristo <kristo@kernel.org>,
	linux-arm-kernel@lists.infradead.org, x86@kernel.org,
	Vinod Koul <vkoul@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Will Deacon <will@kernel.org>, Sinan Kaya <okaya@kernel.org>
Subject: Re: [patch 29/37] PCI/MSI: Use __msi_get_virq() in pci_get_vector()
Date: Sun, 28 Nov 2021 22:00:26 +0100	[thread overview]
Message-ID: <87tufwdmhh.ffs@tglx> (raw)
In-Reply-To: <871r30rrzq.wl-maz@kernel.org>

On Sun, Nov 28 2021 at 19:37, Marc Zyngier wrote:
> On Sat, 27 Nov 2021 01:22:03 +0000,
> Thomas Gleixner <tglx@linutronix.de> wrote:
>
> I worked around it with the hack below, but I doubt this is the real
> thing. portdrv_core.c does complicated things, and I don't completely
> understand its logic.
>
> 	M.
>
> diff --git a/drivers/pci/msi/msi.c b/drivers/pci/msi/msi.c
> index 1f72bc734226..b15278a5fb4b 100644
> --- a/drivers/pci/msi/msi.c
> +++ b/drivers/pci/msi/msi.c
> @@ -1092,8 +1092,9 @@ int pci_irq_vector(struct pci_dev *dev, unsigned int nr)
>  	int irq = __msi_get_virq(&dev->dev, nr);
>  
>  	switch (irq) {
> -	case -ENODEV: return !nr ? dev->irq : -EINVAL;
> -	case -ENOENT: return -EINVAL;
> +	case -ENOENT:
> +	case -ENODEV:
> +		return !nr ? dev->irq : -EINVAL;

Hrm. ENODEV is returned when dev->msi.data == NULL, ENOENT when there is
no MSI entry. But yes, that goes south when the device tried to enable
MSI[X} and then ended up with INTx. It still has dev->msi.data, which
causes it to return -ENOENT, which makes the above go belly up.

Moo, what was I thinking?

Thanks,

        tglx
---
--- a/drivers/pci/msi/msi.c
+++ b/drivers/pci/msi/msi.c
@@ -1032,13 +1032,13 @@ EXPORT_SYMBOL(pci_free_irq_vectors);
  */
 int pci_irq_vector(struct pci_dev *dev, unsigned int nr)
 {
-	int irq = __msi_get_virq(&dev->dev, nr);
+	unsigned int irq;
 
-	switch (irq) {
-	case -ENODEV: return !nr ? dev->irq : -EINVAL;
-	case -ENOENT: return -EINVAL;
-	}
-	return irq;
+	if (!dev->msi_enabled && !dev->msix_enabled)
+		return !nr ? dev->irq : -EINVAL;
+
+	irq = msi_get_virq(&dev->dev, nr);
+	return irq ? irq : -EINVAL;
 }
 EXPORT_SYMBOL(pci_irq_vector);
 
--- a/include/linux/msi.h
+++ b/include/linux/msi.h
@@ -169,21 +169,7 @@ static inline bool msi_device_has_proper
 }
 #endif
 
-int __msi_get_virq(struct device *dev, unsigned int index);
-
-/**
- * msi_get_virq - Return Linux interrupt number of a MSI interrupt
- * @dev:	Device to operate on
- * @index:	MSI interrupt index to look for (0-based)
- *
- * Return: The Linux interrupt number on success (> 0), 0 if not found
- */
-static inline unsigned int msi_get_virq(struct device *dev, unsigned int index)
-{
-	int ret = __msi_get_virq(dev, index);
-
-	return ret < 0 ? 0 : ret;
-}
+unsigned int msi_get_virq(struct device *dev, unsigned int index);
 
 /* Helpers to hide struct msi_desc implementation details */
 #define msi_desc_to_dev(desc)		((desc)->dev)
--- a/kernel/irq/msi.c
+++ b/kernel/irq/msi.c
@@ -119,21 +119,19 @@ int msi_setup_device_data(struct device
 }
 
 /**
- * __msi_get_virq - Return Linux interrupt number of a MSI interrupt
+ * msi_get_virq - Return Linux interrupt number of a MSI interrupt
  * @dev:	Device to operate on
  * @index:	MSI interrupt index to look for (0-based)
  *
- * Return: The Linux interrupt number on success (> 0)
- *	   -ENODEV when the device is not using MSI
- *	   -ENOENT if no such entry exists
+ * Return: The Linux interrupt number on success (> 0), 0 if not found
  */
-int __msi_get_virq(struct device *dev, unsigned int index)
+unsigned int msi_get_virq(struct device *dev, unsigned int index)
 {
 	struct msi_desc *desc;
 	bool pcimsi;
 
 	if (!dev->msi.data)
-		return -ENODEV;
+		return 0;
 
 	pcimsi = msi_device_has_property(dev, MSI_PROP_PCI_MSI);
 
@@ -152,9 +150,9 @@ int __msi_get_virq(struct device *dev, u
 		if (desc->msi_index == index)
 			return desc->irq;
 	}
-	return -ENOENT;
+	return 0;
 }
-EXPORT_SYMBOL_GPL(__msi_get_virq);
+EXPORT_SYMBOL_GPL(msi_get_virq);
 
 #ifdef CONFIG_SYSFS
 static ssize_t msi_mode_show(struct device *dev, struct device_attribute *attr,

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

  reply	other threads:[~2021-11-28 21:01 UTC|newest]

Thread overview: 97+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-11-27  1:20 [patch 00/37] genirq/msi, PCI/MSI: Spring cleaning - Part 2 Thomas Gleixner
2021-11-27  1:20 ` [patch 01/37] device: Move MSI related data into a struct Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27 12:11   ` Greg Kroah-Hartman
2021-11-27  1:20 ` [patch 02/37] device: Add device::msi_data pointer and struct msi_device_data Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27 12:12   ` Greg Kroah-Hartman
2021-11-28  0:14   ` Jason Gunthorpe
2021-11-28 19:09     ` Thomas Gleixner
2021-11-27  1:20 ` [patch 03/37] PCI/MSI: Allocate MSI device data on first use Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 04/37] PCI/MSI: Use lock from msi_device_data Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27 12:13   ` Greg Kroah-Hartman
2021-11-27  1:20 ` [patch 05/37] platform-msi: Allocate MSI device data on first use Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 06/37] bus: fsl-mc-msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 07/37] soc: ti: ti_sci_inta_msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 08/37] genirq/msi: Provide msi_device_populate/destroy_sysfs() Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-30 11:53   ` Jonathan Cameron
2021-11-27  1:20 ` [patch 09/37] PCI/MSI: Let the irq code handle sysfs groups Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 10/37] platform-msi: Let the core " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 11/37] genirq/msi: Remove the original sysfs interfaces Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 12/37] platform-msi: Rename functions and clarify comments Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 13/37] platform-msi: Store platform private data pointer in msi_device_data Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 14/37] genirq/msi: Consolidate MSI descriptor data Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 15/37] platform-msi: Use msi_desc::msi_index Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 16/37] bus: fsl-mc-msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 17/37] soc: ti: ti_sci_inta_msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 18/37] PCI/MSI: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 19/37] genirq/msi: Add msi_device_data::properties Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 20/37] PCI/MSI: Store properties in device::msi::data Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 21/37] x86/pci/XEN: Use device MSI properties Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 22/37] x86/apic/msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 23/37] genirq/msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 24/37] powerpc/cell/axon_msi: Use MSI device properties Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 25/37] powerpc/pseries/msi: " Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 26/37] PCI/MSI: Provide MSI_FLAG_MSIX_CONTIGUOUS Thomas Gleixner
2021-11-27  1:21   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 27/37] powerpc/pseries/msi: Let core code check for contiguous entries Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 28/37] genirq/msi: Provide interface to retrieve Linux interrupt number Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 29/37] PCI/MSI: Use __msi_get_virq() in pci_get_vector() Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-28 19:37   ` Marc Zyngier
2021-11-28 21:00     ` Thomas Gleixner [this message]
2021-11-27  1:20 ` [patch 30/37] PCI/MSI: Simplify pci_irq_get_affinity() Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 31/37] dmaengine: mv_xor_v2: Get rid of msi_desc abuse Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 32/37] perf/smmuv3: Use msi_get_virq() Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:20 ` [patch 33/37] iommu/arm-smmu-v3: " Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-29 10:55   ` Will Deacon
2021-11-29 12:52     ` Thomas Gleixner
2021-11-29 12:58       ` Thomas Gleixner
2021-11-29 13:13     ` Robin Murphy
2021-11-29 14:42       ` Thomas Gleixner
2021-11-29 14:54         ` Robin Murphy
2021-11-30  9:36           ` Will Deacon
2021-11-30 12:30             ` Thomas Gleixner
2021-11-29 13:25   ` Robin Murphy
2021-11-27  1:21 ` [patch 34/37] mailbox: bcm-flexrm-mailbox: Rework MSI interrupt handling Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:21 ` [patch 35/37] bus: fsl-mc: fsl-mc-allocator: Rework MSI handling Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:21 ` [patch 36/37] soc: ti: ti_sci_inta_msi: Get rid of ti_sci_inta_msi_get_virq() Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-27  1:21 ` [patch 37/37] dmaengine: qcom_hidma: Cleanup MSI handling Thomas Gleixner
2021-11-27  1:22   ` Thomas Gleixner
2021-11-29 19:56   ` Sinan Kaya
2021-11-27  1:21 ` [patch 00/37] genirq/msi, PCI/MSI: Spring cleaning - Part 2 Thomas Gleixner
2021-11-27 12:17 ` Greg Kroah-Hartman
2021-11-28  0:39 ` Jason Gunthorpe
2021-11-28 20:27   ` Thomas Gleixner

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=87tufwdmhh.ffs@tglx \
    --to=tglx@linutronix.de \
    --cc=alex.williamson@redhat.com \
    --cc=ashok.raj@intel.com \
    --cc=dmaengine@vger.kernel.org \
    --cc=gregkh@linuxfoundation.org \
    --cc=helgaas@kernel.org \
    --cc=iommu@lists.linux-foundation.org \
    --cc=jgg@nvidia.com \
    --cc=kevin.tian@intel.com \
    --cc=kristo@kernel.org \
    --cc=laurentiu.tudor@nxp.com \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@vger.kernel.org \
    --cc=mark.rutland@arm.com \
    --cc=maz@kernel.org \
    --cc=megha.dey@intel.com \
    --cc=nm@ti.com \
    --cc=okaya@kernel.org \
    --cc=ssantosh@kernel.org \
    --cc=stuyoder@gmail.com \
    --cc=vkoul@kernel.org \
    --cc=will@kernel.org \
    --cc=x86@kernel.org \
    /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).