linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: robin.murphy@arm.com (Robin Murphy)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH v2] ACPI/IORT: Fix PCI ACS enablement
Date: Tue, 3 Oct 2017 14:47:28 +0100	[thread overview]
Message-ID: <121eb6d7-1ac7-ecfa-5ec4-75e5446f5e85@arm.com> (raw)
In-Reply-To: <1506965324-5818-1-git-send-email-lorenzo.pieralisi@arm.com>

On 02/10/17 18:28, Lorenzo Pieralisi wrote:
> commit f6810c15cf97 ("iommu/arm-smmu: Clean up early-probing
> workarounds") removed kernel code that was allowing to initialize
> and probe the SMMU devices early (ie earlier than PCI devices, through
> linker script callback entries) in the boot process because it was not
> needed any longer in that the SMMU devices/drivers now support deferred
> probing.
> 
> Since the SMMUs probe routines are also in charge of requesting global
> PCI ACS kernel enablement, commit f6810c15cf97 ("iommu/arm-smmu: Clean
> up early-probing workarounds") also postponed PCI ACS enablement to
> SMMUs devices probe time, which is too late given that PCI devices needs
> to detect if PCI ACS is enabled to init the respective capability
> through the following call path:
> 
> pci_device_add()
>  -> pci_init_capabilities()
>   -> pci_enable_acs()
> 
> Add code in the ACPI IORT SMMU platform devices initialization path
> (that is called before ACPI PCI enumeration) to detect if there
> exists firmware mappings to map root complexes ids to SMMU ids
> and if so enable ACS for the system.

As a self-contained fix I think this looks reasonable and sufficiently
unintrusive if it wants backporting:

Reviewed-by: Robin Murphy <robin.murphy@arm.com>

More generally, though, I do wonder why this pci_request_acs() mechanism
is like it is - Alex, Bjorn, is there any particular reason that IOMMU
drivers couldn't call pci_enable_acs() directly from their .add_device
callback? In the cases we care about, we can still rely on the IOMMU
probing before any bridge or endpoint drivers, so AFAICS at that point
nothing should have changed much since pci_device_add() time.

Robin.

> Fixes: f6810c15cf97 ("iommu/arm-smmu: Clean up early-probing
> Signed-workarounds")
> Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
> Cc: Will Deacon <will.deacon@arm.com>
> Cc: Hanjun Guo <hanjun.guo@linaro.org>
> Cc: Sudeep Holla <sudeep.holla@arm.com>
> Cc: Catalin Marinas <catalin.marinas@arm.com>
> Cc: Robin Murphy <robin.murphy@arm.com>
> Cc: Zhou Wang <wangzhou1@hisilicon.com>
> Cc: Alex Williamson <alex.williamson@redhat.com>
> ---
> v1 -> v2:
> 
> - Reworked ACS enablement logic and based on root complex to SMMU
>   ids firmware mapping detection
> - Rebased against v4.14-rc3
> 
> v1: https://marc.info/?l=linux-acpi&m=150584059818925&w=2
> 
>  drivers/acpi/arm64/iort.c | 35 +++++++++++++++++++++++++++++++++++
>  1 file changed, 35 insertions(+)
> 
> diff --git a/drivers/acpi/arm64/iort.c b/drivers/acpi/arm64/iort.c
> index 9565d57..de56394 100644
> --- a/drivers/acpi/arm64/iort.c
> +++ b/drivers/acpi/arm64/iort.c
> @@ -1178,12 +1178,44 @@ static int __init iort_add_smmu_platform_device(struct acpi_iort_node *node)
>  	return ret;
>  }
>  
> +static bool __init iort_enable_acs(struct acpi_iort_node *iort_node)
> +{
> +	if (iort_node->type == ACPI_IORT_NODE_PCI_ROOT_COMPLEX) {
> +		struct acpi_iort_node *parent;
> +		struct acpi_iort_id_mapping *map;
> +		int i;
> +
> +		map = ACPI_ADD_PTR(struct acpi_iort_id_mapping, iort_node,
> +				   iort_node->mapping_offset);
> +
> +		for (i = 0; i < iort_node->mapping_count; i++, map++) {
> +			if (!map->output_reference)
> +				continue;
> +
> +			parent = ACPI_ADD_PTR(struct acpi_iort_node,
> +					iort_table,  map->output_reference);
> +			/*
> +			 * If we detect a RC->SMMU mapping, make sure
> +			 * we enable ACS on the system.
> +			 */
> +			if ((parent->type == ACPI_IORT_NODE_SMMU) ||
> +				(parent->type == ACPI_IORT_NODE_SMMU_V3)) {
> +				pci_request_acs();
> +				return true;
> +			}
> +		}
> +	}
> +
> +	return false;
> +}
> +
>  static void __init iort_init_platform_devices(void)
>  {
>  	struct acpi_iort_node *iort_node, *iort_end;
>  	struct acpi_table_iort *iort;
>  	struct fwnode_handle *fwnode;
>  	int i, ret;
> +	bool acs_enabled = false;
>  
>  	/*
>  	 * iort_table and iort both point to the start of IORT table, but
> @@ -1203,6 +1235,9 @@ static void __init iort_init_platform_devices(void)
>  			return;
>  		}
>  
> +		if (!acs_enabled)
> +			acs_enabled = iort_enable_acs(iort_node);
> +
>  		if ((iort_node->type == ACPI_IORT_NODE_SMMU) ||
>  			(iort_node->type == ACPI_IORT_NODE_SMMU_V3)) {
>  
> 

  parent reply	other threads:[~2017-10-03 13:47 UTC|newest]

Thread overview: 8+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-10-02 17:28 [PATCH v2] ACPI/IORT: Fix PCI ACS enablement Lorenzo Pieralisi
2017-10-03 12:48 ` Lorenzo Pieralisi
2017-10-03 19:08   ` Nate Watterson
2017-10-09  8:40   ` Zhou Wang
2017-10-03 13:47 ` Robin Murphy [this message]
2017-10-03 14:45 ` John Garry
2017-10-03 16:53   ` Lorenzo Pieralisi
2017-10-04 16:36 ` Catalin Marinas

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=121eb6d7-1ac7-ecfa-5ec4-75e5446f5e85@arm.com \
    --to=robin.murphy@arm.com \
    --cc=linux-arm-kernel@lists.infradead.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).