public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Laszlo Ersek <lersek@redhat.com>
To: Brijesh Singh <brijesh.singh@amd.com>, edk2-devel@lists.01.org
Cc: Thomas.Lendacky@amd.com, leo.duran@amd.com,
	Jordan Justen <jordan.l.justen@intel.com>
Subject: Re: [RFC v4 11/13] OvmfPkg/QemuFwCfgLib: Implement SEV internal function for Dxe phase
Date: Thu, 11 May 2017 19:07:03 +0200	[thread overview]
Message-ID: <b7739dd6-d682-c2cf-2511-4f9bdfc82446@redhat.com> (raw)
In-Reply-To: <1494454162-9940-12-git-send-email-brijesh.singh@amd.com>

comments below:

On 05/11/17 00:09, Brijesh Singh wrote:
> When SEV is enabled, the DMA must be performed on unencrypted pages.
> So when get asked to perfom FWCFG DMA read or write, we allocate a
> intermediate (bounce buffer) unencrypted buffer and use this buffer
> for DMA read or write.
> 
> 
> Cc: Jordan Justen <jordan.l.justen@intel.com>
> Cc: Laszlo Ersek <lersek@redhat.com>
> Contributed-under: TianoCore Contribution Agreement 1.0
> Signed-off-by: Brijesh Singh <brijesh.singh@amd.com>
> ---
>  OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxeLib.inf |   4 +
>  OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxe.c      | 107 ++++++++++++++++++++
>  2 files changed, 111 insertions(+)
> 
> diff --git a/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxeLib.inf b/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxeLib.inf
> index 346bb881ffc1..f8df77f788b7 100644
> --- a/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxeLib.inf
> +++ b/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxeLib.inf
> @@ -39,6 +39,7 @@ [Sources]
>  
>  [Packages]
>    MdePkg/MdePkg.dec
> +  MdeModulePkg/MdeModulePkg.dec
>    OvmfPkg/OvmfPkg.dec
>  
>  [LibraryClasses]
> @@ -47,4 +48,7 @@ [LibraryClasses]
>    DebugLib
>    IoLib
>    MemoryAllocationLib
> +  MemEncryptSevLib
>  
> +[Protocols]
> +  gEdkiiIoMmuProtocolGuid                         ## SOMETIMES_CONSUMES
> diff --git a/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxe.c b/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxe.c
> index ac05f4c347f3..059666ffa99b 100644
> --- a/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxe.c
> +++ b/OvmfPkg/Library/QemuFwCfgLib/QemuFwCfgDxe.c
> @@ -4,6 +4,7 @@
>  
>    Copyright (C) 2013, Red Hat, Inc.
>    Copyright (c) 2011 - 2013, Intel Corporation. All rights reserved.<BR>
> +  Copyright (c) 2017, Advanced Micro Devices. All rights reserved.<BR>
>  
>    This program and the accompanying materials are licensed and made available
>    under the terms and conditions of the BSD License which accompanies this
> @@ -14,14 +15,36 @@
>    WARRANTIES OR REPRESENTATIONS OF ANY KIND, EITHER EXPRESS OR IMPLIED.
>  **/
>  
> +#include "Uefi.h"

(1) Not sure what this is needed for, but if it is needed, please use
<Uefi.h>.

> +
> +#include <Protocol/IoMmu.h>
> +
> +#include <Library/BaseLib.h>
>  #include <Library/DebugLib.h>
>  #include <Library/QemuFwCfgLib.h>
> +#include <Library/UefiBootServicesTableLib.h>
> +#include <Library/MemEncryptSevLib.h>
>  
>  #include "QemuFwCfgLibInternal.h"
>  
>  STATIC BOOLEAN mQemuFwCfgSupported = FALSE;
>  STATIC BOOLEAN mQemuFwCfgDmaSupported;
>  
> +STATIC EDKII_IOMMU_PROTOCOL        *mIoMmuProtocol;
> +/**
> +
> + Returns a boolean indicating whether SEV is enabled
> +
> + @retval    TRUE    SEV is enabled
> + @retval    FALSE   SEV is disabled
> +**/
> +BOOLEAN
> +InternalQemuFwCfgSevIsEnabled (
> +  VOID
> +  )
> +{
> +  return MemEncryptSevIsEnabled ();
> +}
>  
>  /**
>    Returns a boolean indicating if the firmware configuration interface
> @@ -79,6 +102,21 @@ QemuFwCfgInitialize (
>      mQemuFwCfgDmaSupported = TRUE;
>      DEBUG ((DEBUG_INFO, "QemuFwCfg interface (DMA) is supported.\n"));
>    }
> +
> +  //
> +  // When SEV is enabled, the AmdSevDxe driver should have installed the IoMMU
> +  // protocol which must be used for mapping host buffer to DMA buffer
> +  //

(2) The comment should be stronger. Please state that the IOMMU protocol
is guaranteed to be available, because the driver is listed in the
APRIORI DXE file.

> +  if (mQemuFwCfgDmaSupported && MemEncryptSevIsEnabled ()) {
> +    EFI_STATUS   Status;
> +
> +    Status = gBS->LocateProtocol (&gEdkiiIoMmuProtocolGuid, NULL, (VOID **)&mIoMmuProtocol);
> +    if (EFI_ERROR(Status)) {

(3) Space missing before the paren.

> +      DEBUG ((DEBUG_WARN, "QemuwCfgSevDma: failed to locate IoMmu protocol, disabling DMA support\n"));
> +      mQemuFwCfgDmaSupported = FALSE;
> +    }
> +  }
> +

(4) Similarly to (2), this is too permissive. Please use DEBUG_ERROR,
ASSERT (FALSE), and CpuDeadLoop() here.

We must not silently degrade the fw_cfg interface to port-io, because
some client modules equate the presence of some fw_cfg files -- which
can also be found via port-io -- with DMA and fw_cfg write capability.
The expectation of those modules is correct, and we must not break it.

>    return RETURN_SUCCESS;
>  }
>  
> @@ -114,3 +152,72 @@ InternalQemuFwCfgDmaIsAvailable (
>  {
>    return mQemuFwCfgDmaSupported;
>  }
> +
> +/**
> + Allocate a bounce buffer for SEV DMA.
> +
> +  @param[in]     NumPage  Number of pages.
> +  @param[out]    Buffer   Allocated DMA Buffer pointer
> +
> +**/
> +VOID
> +InternalQemuFwCfgSevDmaAllocateBuffer (
> +  IN     UINT32   NumPages,
> +  OUT    VOID     **Buffer
> +  )
> +{
> +  EFI_STATUS    Status;
> +
> +  if (!mIoMmuProtocol) {
> +    //
> +    // We should never reach here
> +    //
> +    ASSERT (FALSE);
> +    CpuDeadLoop ();
> +  }

(5) In turn, this can be replaced with

  ASSERT (mIoMmuProtocol != NULL);

> +
> +  Status = mIoMmuProtocol->AllocateBuffer (
> +                            mIoMmuProtocol,
> +                            0,
> +                            EfiBootServicesData,
> +                            NumPages,
> +                            Buffer,
> +                            EDKII_IOMMU_ATTRIBUTE_MEMORY_CACHED
> +                          );
> +  ASSERT_EFI_ERROR (Status);

(6) Please add a CpuDeadLoop() here.

> +
> +  DEBUG ((DEBUG_VERBOSE, "QemuFwCfgSevDma allocate buffer 0x%Lx Pages %d\n", (UINTN)Buffer, NumPages));
> +
> +}

(7) Again I suggest to add gEfiCallerBaseName and __FUNCTION__ to the
debug message.

(8) Please also wrap it to 79 chars.

(9) %Lx is not right for printing a UINTN. There is no conversion
specifier that directly matches UINTN, so you'll have to use

  (UINT64)(UINTN)Buffer

for portability between IA32 and X64.

(10) NumPages should be printed with %u or %x, it is unsigned.

> +
> +/**
> + Free the DMA buffer allocated using InternalQemuFwCfgSevDmaAllocateBuffer
> +
> +  @param[in]     NumPage  Number of pages.
> +  @param[in]     Buffer   DMA Buffer pointer
> +
> +**/
> +VOID
> +InternalQemuFwCfgSevDmaFreeBuffer (
> +  IN     VOID     *Buffer,
> +  IN     UINT32   NumPages
> +  )
> +{
> +  EFI_STATUS    Status;
> +
> +  if (!mIoMmuProtocol) {
> +    //
> +    // We should never reach here
> +    //
> +    ASSERT (FALSE);
> +    CpuDeadLoop ();
> +  }
> +
> +  Status = mIoMmuProtocol->FreeBuffer (
> +                            mIoMmuProtocol,
> +                            NumPages,
> +                            Buffer
> +                          );
> +  ASSERT_EFI_ERROR (Status);
> +  DEBUG ((DEBUG_VERBOSE, "QemuFwCfgSevDma free buffer 0x%Lx Pages %d\n", (UINTN)Buffer, NumPages));
> +}
> 

The same comments apply here.

Thanks!
Laszlo


  reply	other threads:[~2017-05-11 17:07 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-10 22:09 [RFC v4 00/13] x86: Secure Encrypted Virtualization (AMD) Brijesh Singh
2017-05-10 22:09 ` [RFC v4 01/13] UefiCpuPkg: Define AMD Memory Encryption specific CPUID and MSR Brijesh Singh
2017-05-11  0:30   ` Fan, Jeff
2017-05-10 22:09 ` [RFC v4 02/13] OvmfPkg/ResetVector: Set C-bit when building initial page table Brijesh Singh
2017-05-11 11:40   ` Laszlo Ersek
2017-05-10 22:09 ` [RFC v4 03/13] OvmfPkg: Update dsc to use IoLib from BaseIoLibIntrinsicSev.inf Brijesh Singh
2017-05-11 11:46   ` Laszlo Ersek
2017-05-10 22:09 ` [RFC v4 04/13] OvmfPkg/BaseMemcryptSevLib: Add SEV helper library Brijesh Singh
2017-05-11 14:04   ` Laszlo Ersek
2017-05-11 18:03     ` Brijesh Singh
2017-05-10 22:09 ` [RFC v4 05/13] OvmfPkg/PlatformPei: Set memory encryption PCD when SEV is enabled Brijesh Singh
2017-05-11 14:37   ` Laszlo Ersek
2017-05-11 18:04     ` Brijesh Singh
2017-05-10 22:09 ` [RFC v4 06/13] OvmfPkg:AmdSevDxe: add AmdSevDxe driver Brijesh Singh
2017-05-11  0:56   ` Yao, Jiewen
2017-05-11 15:19   ` Laszlo Ersek
2017-05-11 15:53     ` Laszlo Ersek
2017-05-11 17:43       ` Jordan Justen
2017-05-11 18:01         ` Brijesh Singh
2017-05-15 17:47           ` Jordan Justen
2017-05-16 12:04             ` Brijesh Singh
2017-05-16 17:56               ` Jordan Justen
2017-05-16 20:25                 ` Brijesh Singh
2017-05-18  8:50                   ` Laszlo Ersek
2017-05-11 20:14         ` Laszlo Ersek
2017-05-11 18:12     ` Brijesh Singh
2017-05-10 22:09 ` [RFC v4 07/13] OvmfPkg/QemuFwCfgLib: Provide Pei and Dxe specific library Brijesh Singh
2017-05-11 15:40   ` Laszlo Ersek
2017-05-11 18:16     ` Brijesh Singh
2017-05-10 22:09 ` [RFC v4 08/13] OvmfPkg/QemuFwCfgLib: Prepare for SEV support Brijesh Singh
2017-05-11 15:57   ` Laszlo Ersek
2017-05-10 22:09 ` [RFC v4 09/13] OvmfPkg/QemuFwCfgLib: Implement SEV internal function for SEC phase Brijesh Singh
2017-05-11 16:24   ` Laszlo Ersek
2017-05-11 18:21     ` Brijesh Singh
2017-05-10 22:09 ` [RFC v4 10/13] OvmfPkg/QemuFwCfgLib: Implement SEV internal functions for PEI phase Brijesh Singh
2017-05-11 16:38   ` Laszlo Ersek
2017-05-10 22:09 ` [RFC v4 11/13] OvmfPkg/QemuFwCfgLib: Implement SEV internal function for Dxe phase Brijesh Singh
2017-05-11 17:07   ` Laszlo Ersek [this message]
2017-05-10 22:09 ` [RFC v4 12/13] OvmfPkg/QemuFwCfgLib: Add option to dynamic alloc FW_CFG_DMA Access Brijesh Singh
2017-05-11 17:10   ` Laszlo Ersek
2017-05-10 22:09 ` [RFC v4 13/13] OvmfPkg/QemuFwCfgLib: Add SEV support Brijesh Singh
2017-05-11 17:44   ` Laszlo Ersek

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-list from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=b7739dd6-d682-c2cf-2511-4f9bdfc82446@redhat.com \
    --to=devel@edk2.groups.io \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox