efi_memattr_init() added a sanity check to avoid firmware caused corruption. The check is based on efi memmap entry numbers, but kexec only takes the runtime related memmap entries thus this caused many false warnings, see below thread for details:
https://lore.kernel.org/all/[email protected]/ Ard suggests to skip the efi memattr table in kexec, this makes sense because those memattr fixups are not critical. Reported-by: Breno Leitao <[email protected]> Reported-and-tested-by: Usama Arif <[email protected]> Suggested-by: Ard Biesheuvel <[email protected]> Signed-off-by: Dave Young <[email protected]> diff --git a/arch/x86/platform/efi/quirks.c b/arch/x86/platform/efi/quirks.c index 846bf49f2508..553f330198f2 100644 --- a/arch/x86/platform/efi/quirks.c +++ b/arch/x86/platform/efi/quirks.c @@ -561,6 +561,11 @@ int __init efi_reuse_config(u64 tables, int nr_tables) if (!efi_guidcmp(guid, SMBIOS_TABLE_GUID)) ((efi_config_table_64_t *)p)->table = data->smbios; + + /* Do not bother to play with mem attr table across kexec */ + if (!efi_guidcmp(guid, EFI_MEMORY_ATTRIBUTES_TABLE_GUID)) + ((efi_config_table_64_t *)p)->table = EFI_INVALID_TABLE_ADDR; + p += sz; } early_memunmap(tablep, nr_tables * sz);
