[RFC][PATCH] arm64: efi: Obey EFI memory type in dmi_remap

Ard Biesheuvel ard.biesheuvel at linaro.org
Sat Jan 17 00:24:15 PST 2015


On 17 January 2015 at 02:24, Laura Abbott <lauraa at codeaurora.org> wrote:
> Currently, dmi_remap unconditionally calls ioremap_cache for dmi_remap.
> The memory that's being remapped may be part of the existing EFI
> memory map and already remapped as uncached. Remapping as cached can
> created unexpected behavior so check the physical address against the
> EFI memory type before remapping.
>

Mapping the SMBIOS tables as uncached is a bad idea, as 'uncached'
implies MT_DEVICE_nGnRnE in the arm64 kernel, and the SMBIOS structure
tables are not guaranteed to be aligned. So you should at least use
MT_NORMAL_NC here.

Why does it end up mapped uncached in the first place? Is it backed by
a ROM instead of normal RAM? And does the region have any of the WT/WC
attributes set in addition to UC? What type is it? I suppose it has
the EFI_MEMORY_RUNTIME bit set as well? (Note that the spec seems to
assume that configuration tables always reside in system RAM [UEFI
spec 2.4A 2.3.6])

Also, with my latest virtmap changes, UEFI runtime regions are only
mapped during runtime service invocations, and explicitly when e.g.
the SMBIOS or ACPI layer needs to get at the data. Currently, regions
backed by system RAM are still covered by the linear mapping as well,
but that is something we intend to change too. So if the region is not
system RAM (!EFI_MEMORY_WB), the latest code will not have this region
mapped at DMI scan time, afaict.

Finally, I have been looking into classifying memory as system RAM or
not myself[0]. Even if SMBIOS is UEFI only in arm64, I would prefer
not to add infrastructure that allows us to answer these kind of
questions only when booted via UEFI.

[0] http://article.gmane.org/gmane.linux.kernel.efi/5133


Regards,
Ard.


> Signed-off-by: Laura Abbott <lauraa at codeaurora.org>
> ---
> Actual bug, not just theoretical problem. I'm marking this as RFC
> because I'm not sure what any kind of specificaiton may have to say.
> ---
>  arch/arm64/include/asm/dmi.h |  5 +++--
>  arch/arm64/kernel/efi.c      | 23 +++++++++++++++++++++++
>  2 files changed, 26 insertions(+), 2 deletions(-)
>
> diff --git a/arch/arm64/include/asm/dmi.h b/arch/arm64/include/asm/dmi.h
> index 69d37d8..f0cc8d0 100644
> --- a/arch/arm64/include/asm/dmi.h
> +++ b/arch/arm64/include/asm/dmi.h
> @@ -22,9 +22,10 @@
>   * request a virtual mapping for configuration tables such as SMBIOS.
>   * This means we have to map them before use.
>   */
> -#define dmi_early_remap(x, l)          ioremap_cache(x, l)
> +#define dmi_early_remap(x, l)          dmi_remap(x, l)
>  #define dmi_early_unmap(x, l)          iounmap(x)
> -#define dmi_remap(x, l)                        ioremap_cache(x, l)
> +
> +extern void __iomem *dmi_remap(phys_addr_t phys_addr, size_t size);
>  #define dmi_unmap(x)                   iounmap(x)
>  #define dmi_alloc(l)                   kzalloc(l, GFP_KERNEL)
>
> diff --git a/arch/arm64/kernel/efi.c b/arch/arm64/kernel/efi.c
> index 82f27c3..139666e 100644
> --- a/arch/arm64/kernel/efi.c
> +++ b/arch/arm64/kernel/efi.c
> @@ -22,6 +22,7 @@
>  #include <linux/slab.h>
>
>  #include <asm/cacheflush.h>
> +#include <asm/io.h>
>  #include <asm/efi.h>
>  #include <asm/tlbflush.h>
>  #include <asm/mmu_context.h>
> @@ -321,6 +322,28 @@ void __init efi_init(void)
>         reserve_regions();
>  }
>
> +static bool is_phys_addr_normal_ram(unsigned long phys_addr)
> +{
> +       efi_memory_desc_t *md;
> +
> +       for_each_efi_memory_desc(&memmap, md) {
> +               if ((md->phys_addr <= phys_addr) &&
> +                   (phys_addr < (md->phys_addr +
> +                                 (md->num_pages << EFI_PAGE_SHIFT))))
> +                       return is_normal_ram(md);
> +       }
> +       return false;
> +}
> +
> +void __iomem *dmi_remap(phys_addr_t phys_addr, size_t size)
> +{
> +       if (is_phys_addr_normal_ram(phys_addr))
> +               return ioremap_cache(phys_addr, size);
> +       else
> +               return ioremap(phys_addr, size);
> +}
> +
> +
>  void __init efi_idmap_init(void)
>  {
>         if (!efi_enabled(EFI_BOOT))
> --
> Qualcomm Innovation Center, Inc.
> Qualcomm Innovation Center, Inc. is a member of Code Aurora Forum, a Linux Foundation Collaborative Project
>



More information about the linux-arm-kernel mailing list