Re: [PATCH v6] x86/PCI: Ignore E820 reservations for bridge windows on newer systems

[Hans de Goede]

Hi All,

On 12/17/21 15:13, Hans de Goede wrote:
> Some BIOS-es contain a bug where they add addresses which map to system
> RAM in the PCI host bridge window returned by the ACPI _CRS method, see
> commit 4dc2287c1805 ("x86: avoid E820 regions when allocating address
> space").
> 
> To work around this bug Linux excludes E820 reserved addresses when
> allocating addresses from the PCI host bridge window since 2010.
> 
> Recently (2019) some systems have shown-up with E820 reservations which
> cover the entire _CRS returned PCI bridge memory window, causing all
> attempts to assign memory to PCI BARs which have not been setup by the
> BIOS to fail. For example here are the relevant dmesg bits from a
> Lenovo IdeaPad 3 15IIL 81WE:
> 
>  [mem 0x000000004bc50000-0x00000000cfffffff] reserved
>  pci_bus 0000:00: root bus resource [mem 0x65400000-0xbfffffff window]
> 
> The ACPI specifications appear to allow this new behavior:
> 
> The relationship between E820 and ACPI _CRS is not really very clear.
> ACPI v6.3, sec 15, table 15-374, says AddressRangeReserved means:
> 
>   This range of addresses is in use or reserved by the system and is
>   not to be included in the allocatable memory pool of the operating
>   system's memory manager.
> 
> and it may be used when:
> 
>   The address range is in use by a memory-mapped system device.
> 
> Furthermore, sec 15.2 says:
> 
>   Address ranges defined for baseboard memory-mapped I/O devices, such
>   as APICs, are returned as reserved.
> 
> A PCI host bridge qualifies as a baseboard memory-mapped I/O device,
> and its apertures are in use and certainly should not be included in
> the general allocatable pool, so the fact that some BIOS-es reports
> the PCI aperture as "reserved" in E820 doesn't seem like a BIOS bug.
> 
> So it seems that the excluding of E820 reserved addresses is a mistake.
> 
> Ideally Linux would fully stop excluding E820 reserved addresses,
> but then the old systems this was added for will regress.
> Instead keep the old behavior for old systems, while ignoring
> the E820 reservations for any systems from now on.
> 
> Old systems are defined here as BIOS year < 2018, this was chosen to make
> sure that E820 reservations will not be used on the currently affected
> systems, while at the same time also taking into account that the systems
> for which the E820 checking was originally added may have received BIOS
> updates for quite a while (esp. CVE related ones), giving them a more
> recent BIOS year then 2010.
> 
> BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=206459
> BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1868899
> BugLink: https://bugzilla.redhat.com/show_bug.cgi?id=1871793
> BugLink: https://bugs.launchpad.net/bugs/1878279
> BugLink: https://bugs.launchpad.net/bugs/1931715
> BugLink: https://bugs.launchpad.net/bugs/1932069
> BugLink: https://bugs.launchpad.net/bugs/1921649
> Cc: Benoit Grégoire <benoitg@xxxxxxxx>
> Cc: Hui Wang <hui.wang@xxxxxxxxxxxxx>
> Cc: stable@xxxxxxxxxxxxxxx
> Reviewed-by: Mika Westerberg <mika.westerberg@xxxxxxxxxxxxxxx>
> Acked-by: Rafael J. Wysocki <rafael.j.wysocki@xxxxxxxxx>
> Acked-by: Bjorn Helgaas <bhelgaas@xxxxxxxxxx>
> Signed-off-by: Hans de Goede <hdegoede@xxxxxxxxxx>
> ---
> Changes in v6:
> - Remove the possibility to change the behavior from the commandline
>   because of worries that users may use this to paper over other problems

ping ?

Regards,

Hans







> Changes in v5:
> - Drop mention of Windows behavior from the commit msg, replace with a
>   reference to the specs
> - Improve documentation in Documentation/admin-guide/kernel-parameters.txt
> - Reword the big comment added, use "PCI host bridge window" in it and drop
>   all refences to Windows
> 
> Changes in v4:
> - Rewrap the big comment block to fit in 80 columns
> - Add Rafael's Acked-by
> - Add Cc: stable@xxxxxxxxxxxxxxx
> 
> Changes in v3:
> - Commit msg tweaks (drop dmesg timestamps, typo fix)
> - Use "defined(CONFIG_...)" instead of "defined CONFIG_..."
> - Add Mika's Reviewed-by
> 
> Changes in v2:
> - Replace the per model DMI quirk approach with disabling E820 reservations
>   checking for all systems with a BIOS year >= 2018
> - Add documentation for the new kernel-parameters to
>   Documentation/admin-guide/kernel-parameters.txt
> ---
> Other patches trying to address the same issue:
> https://lore.kernel.org/r/20210624095324.34906-1-hui.wang@xxxxxxxxxxxxx
> https://lore.kernel.org/r/20200617164734.84845-1-mika.westerberg@xxxxxxxxxxxxxxx
> V1 patch:
> https://lore.kernel.org/r/20211005150956.303707-1-hdegoede@xxxxxxxxxx
> ---
>  arch/x86/kernel/resource.c | 23 ++++++++++++++++++++++-
>  1 file changed, 22 insertions(+), 1 deletion(-)
> 
> diff --git a/arch/x86/kernel/resource.c b/arch/x86/kernel/resource.c
> index 9b9fb7882c20..9ae64f9af956 100644
> --- a/arch/x86/kernel/resource.c
> +++ b/arch/x86/kernel/resource.c
> @@ -1,4 +1,5 @@
>  // SPDX-License-Identifier: GPL-2.0
> +#include <linux/dmi.h>
>  #include <linux/ioport.h>
>  #include <asm/e820/api.h>
>  
> @@ -23,11 +24,31 @@ static void resource_clip(struct resource *res, resource_size_t start,
>  		res->start = end + 1;
>  }
>  
> +/*
> + * Some BIOS-es contain a bug where they add addresses which map to
> + * system RAM in the PCI host bridge window returned by the ACPI _CRS
> + * method, see commit 4dc2287c1805 ("x86: avoid E820 regions when
> + * allocating address space"). To avoid this Linux by default excludes
> + * E820 reservations when allocating addresses since 2010.
> + * In 2019 some systems have shown-up with E820 reservations which cover
> + * the entire _CRS returned PCI host bridge window, causing all attempts
> + * to assign memory to PCI BARs to fail if Linux uses E820 reservations.
> + *
> + * Ideally Linux would fully stop using E820 reservations, but then
> + * the old systems this was added for will regress.
> + * Instead keep the old behavior for old systems, while ignoring the
> + * E820 reservations for any systems from now on.
> + */
>  static void remove_e820_regions(struct resource *avail)
>  {
> -	int i;
> +	int i, year = dmi_get_bios_year();
>  	struct e820_entry *entry;
>  
> +	if (year >= 2018)
> +		return;
> +
> +	pr_info_once("PCI: Removing E820 reservations from host bridge windows\n");
> +
>  	for (i = 0; i < e820_table->nr_entries; i++) {
>  		entry = &e820_table->entries[i];
>  
>