linux-kernel - Re: [PATCHv4] PCI/ACPI: _DSM PRESERVE_BOOT_CONFIG function rev id doesn't match with spec

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <20250227184053.GA609152@bhelgaas>
Date: Thu, 27 Feb 2025 12:40:53 -0600
From: Bjorn Helgaas <helgaas@...nel.org>
To: Zhou Shengqing <zhoushengqing@...info.com>
Cc: Bjorn Helgaas <bhelgaas@...gle.com>,
	"Rafael J. Wysocki" <rafael@...nel.org>,
	Len Brown <lenb@...nel.org>, linux-acpi@...r.kernel.org,
	linux-pci@...r.kernel.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCHv4] PCI/ACPI: _DSM PRESERVE_BOOT_CONFIG function rev id
 doesn't match with spec

On Mon, Dec 16, 2024 at 05:27:51AM +0000, Zhou Shengqing wrote:
> Per PCI Firmware Specification Revision 3.3 Table 4-7 _DSM Definitions
> for PCI. Preserve PCI Boot Configuration Initial Revision ID changed to 2.
> But the code remains unchanged, still 1.
> 
> v4:Initialize *obj to NULL.
> v3:try revision id 1 first, then try revision id 2.
> v2:add Fixes tag.

Thanks for working on this issue.

  - Thanks for the revision history.  For future posts, put it below
    the "---" line so it's in the email but not part of the commit log.

  - I think there's a leak in pci_acpi_preserve_config() because it
    doesn't free "obj" before it returns true.  If you agree, add a
    preparatory patch to fix this.

  - Add a preparatory patch to return false early in
    pci_acpi_preserve_config() if !ACPI_HANDLE(&host_bridge->dev) so
    the body of the function is unindented, similar to what
    acpi_pci_add_bus() does.

  - Add another preparatory patch that adds acpi_check_dsm() of the
    desired function/rev ID for DSM_PCI_PRESERVE_BOOT_CONFIG,
    DSM_PCI_POWER_ON_RESET_DELAY, DSM_PCI_DEVICE_READINESS_DURATIONS.
    Move the "Evaluate PCI Boot Configuration" comment above the
    acpi_check_dsm() since it applies to the whole function, not just
    the rev 1 code in this patch.

  - Rework this patch so it only adds acpi_check_dsm() and
    acpi_evaluate_dsm_typed() for rev 2.

  - Throughout, wrap code and comments to fit in 80 columns because
    that's the convention for the rest of the file.

  - You can use "BIT(DSM_PCI_PRESERVE_BOOT_CONFIG)" and similar
    instead of "1ULL << DSM_PCI_PRESERVE_BOOT_CONFIG" when calling
    acpi_check_dsm().

  - Add a [0/n] cover letter when posting the series.  Each patch
    should be a response to the cover letter.  "git format-patch" and
    "git send-email" will do that for you automatically.

> Fixes: 9d7d5db8e78e ("PCI: Move PRESERVE_BOOT_CONFIG _DSM evaluation to pci_register_host_bridge()")
> 
> Signed-off-by: Zhou Shengqing <zhoushengqing@...info.com>
> ---
>  drivers/pci/pci-acpi.c | 42 ++++++++++++++++++++++++++++++++----------
>  1 file changed, 32 insertions(+), 10 deletions(-)
> 
> diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
> index af370628e583..f805cd134019 100644
> --- a/drivers/pci/pci-acpi.c
> +++ b/drivers/pci/pci-acpi.c
> @@ -123,19 +123,41 @@ phys_addr_t acpi_pci_root_get_mcfg_addr(acpi_handle handle)
>  bool pci_acpi_preserve_config(struct pci_host_bridge *host_bridge)
>  {
>  	if (ACPI_HANDLE(&host_bridge->dev)) {
> -		union acpi_object *obj;
> +		union acpi_object *obj = NULL;
>  
>  		/*
> -		 * Evaluate the "PCI Boot Configuration" _DSM Function.  If it
> -		 * exists and returns 0, we must preserve any PCI resource
> -		 * assignments made by firmware for this host bridge.
> +		 * Per PCI Firmware r3.2, released Jan 26, 2015,
> +		 * DSM_PCI_PRESERVE_BOOT_CONFIG Revision ID is 1.
> +		 * But PCI Firmware r3.3, released Jan 20, 2021,
> +		 * changed sec 4.6.5 to say
> +		 * "lowest valid Revision ID value: 2". So try revision 1
> +		 * first for old platform, then try revision 2.
>  		 */
> -		obj = acpi_evaluate_dsm_typed(ACPI_HANDLE(&host_bridge->dev),
> -					      &pci_acpi_dsm_guid,
> -					      1, DSM_PCI_PRESERVE_BOOT_CONFIG,
> -					      NULL, ACPI_TYPE_INTEGER);
> -		if (obj && obj->integer.value == 0)
> -			return true;
> +		if (acpi_check_dsm(ACPI_HANDLE(&host_bridge->dev), &pci_acpi_dsm_guid, 1,
> +				   1ULL << DSM_PCI_PRESERVE_BOOT_CONFIG)) {
> +			/*
> +			 * Evaluate the "PCI Boot Configuration" _DSM Function.  If it
> +			 * exists and returns 0, we must preserve any PCI resource
> +			 * assignments made by firmware for this host bridge.
> +			 */
> +			obj = acpi_evaluate_dsm_typed(ACPI_HANDLE(&host_bridge->dev),
> +						      &pci_acpi_dsm_guid,
> +						      1, DSM_PCI_PRESERVE_BOOT_CONFIG,
> +						      NULL, ACPI_TYPE_INTEGER);
> +			if (obj && obj->integer.value == 0)
> +				return true;
> +		}
> +
> +		if (acpi_check_dsm(ACPI_HANDLE(&host_bridge->dev), &pci_acpi_dsm_guid, 2,
> +				   1ULL << DSM_PCI_PRESERVE_BOOT_CONFIG)) {
> +			obj = acpi_evaluate_dsm_typed(ACPI_HANDLE(&host_bridge->dev),
> +						      &pci_acpi_dsm_guid,
> +						      2, DSM_PCI_PRESERVE_BOOT_CONFIG,
> +						      NULL, ACPI_TYPE_INTEGER);
> +			if (obj && obj->integer.value == 0)
> +				return true;
> +		}
> +
>  		ACPI_FREE(obj);
>  	}
>  
> -- 
> 2.39.2
>