linux-kernel - Re: [RFC PATCH v4 10/28] x86: Insure that boot memory areas are mapped properly

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] [thread-next>] [day] [month] [year] [list]
Message-ID: <7e9fa3ee-b120-998c-9752-53f7bc3d1d0f@amd.com>
Date:   Wed, 22 Feb 2017 12:34:39 -0600
From:   Tom Lendacky <thomas.lendacky@....com>
To:     Borislav Petkov <bp@...en8.de>
CC:     <linux-arch@...r.kernel.org>, <linux-efi@...r.kernel.org>,
        <kvm@...r.kernel.org>, <linux-doc@...r.kernel.org>,
        <x86@...nel.org>, <linux-kernel@...r.kernel.org>,
        <kasan-dev@...glegroups.com>, <linux-mm@...ck.org>,
        <iommu@...ts.linux-foundation.org>, Rik van Riel <riel@...hat.com>,
        Radim Krčmář <rkrcmar@...hat.com>,
        Toshimitsu Kani <toshi.kani@....com>,
        Arnd Bergmann <arnd@...db.de>,
        Jonathan Corbet <corbet@....net>,
        Matt Fleming <matt@...eblueprint.co.uk>,
        "Michael S. Tsirkin" <mst@...hat.com>,
        Joerg Roedel <joro@...tes.org>,
        Konrad Rzeszutek Wilk <konrad.wilk@...cle.com>,
        Paolo Bonzini <pbonzini@...hat.com>,
        Brijesh Singh <brijesh.singh@....com>,
        Ingo Molnar <mingo@...hat.com>,
        Alexander Potapenko <glider@...gle.com>,
        Andy Lutomirski <luto@...nel.org>,
        "H. Peter Anvin" <hpa@...or.com>,
        Andrey Ryabinin <aryabinin@...tuozzo.com>,
        Thomas Gleixner <tglx@...utronix.de>,
        Larry Woodman <lwoodman@...hat.com>,
        Dmitry Vyukov <dvyukov@...gle.com>
Subject: Re: [RFC PATCH v4 10/28] x86: Insure that boot memory areas are
 mapped properly

On 2/20/2017 1:45 PM, Borislav Petkov wrote:
> On Thu, Feb 16, 2017 at 09:44:11AM -0600, Tom Lendacky wrote:
>> The boot data and command line data are present in memory in a decrypted
>> state and are copied early in the boot process.  The early page fault
>> support will map these areas as encrypted, so before attempting to copy
>> them, add decrypted mappings so the data is accessed properly when copied.
>>
>> For the initrd, encrypt this data in place. Since the future mapping of the
>> initrd area will be mapped as encrypted the data will be accessed properly.
>>
>> Signed-off-by: Tom Lendacky <thomas.lendacky@....com>
>> ---
>
> ...
>
>> diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
>> index 182a4c7..03f8e74 100644
>> --- a/arch/x86/kernel/head64.c
>> +++ b/arch/x86/kernel/head64.c
>> @@ -46,13 +46,18 @@ static void __init reset_early_page_tables(void)
>>  	write_cr3(__sme_pa_nodebug(early_level4_pgt));
>>  }
>>
>> +void __init __early_pgtable_flush(void)
>> +{
>> +	write_cr3(__sme_pa_nodebug(early_level4_pgt));
>> +}
>
> Move that to mem_encrypt.c where it is used and make it static. The diff
> below, ontop of this patch, seems to build fine here.

Ok, I can do that.

>
> Also, aren't those mappings global so that you need to toggle CR4.PGE
> for that?
>
> PAGE_KERNEL at least has _PAGE_GLOBAL set.

The early_pmd_flags has _PAGE_GLOBAL cleared:

pmdval_t early_pmd_flags = __PAGE_KERNEL_LARGE & ~(_PAGE_GLOBAL | _PAGE_NX);

so I didn't do the CR4.PGE toggle. I could always add it to be safe in
case that is ever changed. It only happens twice, on the map and on the
unmap, so it shouldn't be a big deal.

>
>> +
>>  /* Create a new PMD entry */
>> -int __init early_make_pgtable(unsigned long address)
>> +int __init __early_make_pgtable(unsigned long address, pmdval_t pmd)
>
> __early_make_pmd() then, since it creates a PMD entry.
>
>>  	unsigned long physaddr = address - __PAGE_OFFSET;
>>  	pgdval_t pgd, *pgd_p;
>>  	pudval_t pud, *pud_p;
>> -	pmdval_t pmd, *pmd_p;
>> +	pmdval_t *pmd_p;
>>
>>  	/* Invalid address or early pgt is done ?  */
>>  	if (physaddr >= MAXMEM || read_cr3() != __sme_pa_nodebug(early_level4_pgt))
>
> ...
>
>> diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
>> index ac3565c..ec548e9 100644
>> --- a/arch/x86/mm/mem_encrypt.c
>> +++ b/arch/x86/mm/mem_encrypt.c
>> @@ -16,8 +16,12 @@
>>
>>  #include <asm/tlbflush.h>
>>  #include <asm/fixmap.h>
>> +#include <asm/setup.h>
>> +#include <asm/bootparam.h>
>>
>>  extern pmdval_t early_pmd_flags;
>> +int __init __early_make_pgtable(unsigned long, pmdval_t);
>> +void __init __early_pgtable_flush(void);
>
> What's with the forward declarations?
>
> Those should be in some header AFAICT.

I can add them to a header, probably arch/x86/include/asm/pgtable.h.

Thanks,
Tom

>
>>   * Since SME related variables are set early in the boot process they must
>> @@ -103,6 +107,76 @@ void __init sme_early_decrypt(resource_size_t paddr, unsigned long size)
>>  	__sme_early_enc_dec(paddr, size, false);
>>  }
>
> ...
>
> ---
> diff --git a/arch/x86/kernel/head64.c b/arch/x86/kernel/head64.c
> index 03f8e74c7223..c47500d72330 100644
> --- a/arch/x86/kernel/head64.c
> +++ b/arch/x86/kernel/head64.c
> @@ -46,11 +46,6 @@ static void __init reset_early_page_tables(void)
>  	write_cr3(__sme_pa_nodebug(early_level4_pgt));
>  }
>
> -void __init __early_pgtable_flush(void)
> -{
> -	write_cr3(__sme_pa_nodebug(early_level4_pgt));
> -}
> -
>  /* Create a new PMD entry */
>  int __init __early_make_pgtable(unsigned long address, pmdval_t pmd)
>  {
> diff --git a/arch/x86/mm/mem_encrypt.c b/arch/x86/mm/mem_encrypt.c
> index ec548e9a76f1..0af020b36232 100644
> --- a/arch/x86/mm/mem_encrypt.c
> +++ b/arch/x86/mm/mem_encrypt.c
> @@ -21,7 +21,7 @@
>
>  extern pmdval_t early_pmd_flags;
>  int __init __early_make_pgtable(unsigned long, pmdval_t);
> -void __init __early_pgtable_flush(void);
> +extern pgd_t early_level4_pgt[PTRS_PER_PGD];
>
>  /*
>   * Since SME related variables are set early in the boot process they must
> @@ -34,6 +34,11 @@ EXPORT_SYMBOL_GPL(sme_me_mask);
>  /* Buffer used for early in-place encryption by BSP, no locking needed */
>  static char sme_early_buffer[PAGE_SIZE] __aligned(PAGE_SIZE);
>
> +static void __init early_pgtable_flush(void)
> +{
> +	write_cr3(__sme_pa_nodebug(early_level4_pgt));
> +}
> +
>  /*
>   * This routine does not change the underlying encryption setting of the
>   * page(s) that map this memory. It assumes that eventually the memory is
> @@ -158,7 +163,7 @@ void __init sme_unmap_bootdata(char *real_mode_data)
>  	 */
>  	__sme_map_unmap_bootdata(real_mode_data, false);
>
> -	__early_pgtable_flush();
> +	early_pgtable_flush();
>  }
>
>  void __init sme_map_bootdata(char *real_mode_data)
> @@ -174,7 +179,7 @@ void __init sme_map_bootdata(char *real_mode_data)
>  	 */
>  	__sme_map_unmap_bootdata(real_mode_data, true);
>
> -	__early_pgtable_flush();
> +	early_pgtable_flush();
>  }
>
>  void __init sme_early_init(void)
>