public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Laszlo Ersek <lersek@redhat.com>
To: Jordan Justen <jordan.l.justen@intel.com>,
	edk2-devel-01 <edk2-devel@lists.01.org>
Cc: Gary Ching-Pang Lin <glin@suse.com>
Subject: Re: [PATCH v2 4/5] OvmfPkg: raise max variable size (auth & non-auth) to 33KB for FD_SIZE_4MB
Date: Thu, 4 May 2017 21:27:33 +0200	[thread overview]
Message-ID: <c6f4983e-be93-93ac-4d0e-e1ce42e43f66@redhat.com> (raw)
In-Reply-To: <149392380410.9783.3695261429662975796@jljusten-skl.jf.intel.com>

On 05/04/17 20:50, Jordan Justen wrote:
> On 2017-05-04 09:52:48, Laszlo Ersek wrote:
>> On 05/04/17 18:36, Jordan Justen wrote:
>>> On 2017-05-03 14:39:46, Laszlo Ersek wrote:
>>>> The "ConfirmSetOfLargeVariable" test case of the Secure Boot Logo Test
>>>> ("Microsoft.UefiSecureBootLogo.Tests") suite in the Microsoft Hardware
>>>> Certification Kit sets a 32 KB large non-authenticated variable.
>>>
>>> According to
>>> http://www.uefi.org/sites/default/files/resources/UEFI_Plugfest_Security_Microsoft_Fall_2016.pdf
>>>
>>> "The maximum supported variable size must be at least 64kB"
>>>
>>> Should we just bump the size to match this? We should be able to make
>>> this change later once it is in a test/spec, but for some reason I
>>> thought the requirement was already 64k.
>>
>> The 32KB requirement comes from the most recent Secure Boot Logo Test. I
>
> If the limit is 32k, why go with 33k? Does the test fail with a 32k
> limit?

Yes, it does. First I tried setting the PCDs to 0x8000 sharp, and then
the test failed.

Then I remembered that the PCDs present a limit on data, name, and
attributes together:

(1) In "MdeModulePkg/Universal/Variable/RuntimeDxe/Variable.c", function
InitNonVolatileVariableStore(), we stash the PCDs like this:

>   mVariableModuleGlobal->MaxVariableSize = PcdGet32 (PcdMaxVariableSize);
>   mVariableModuleGlobal->MaxAuthVariableSize = ((PcdGet32 (PcdMaxAuthVariableSize) != 0) ? PcdGet32 (PcdMaxAuthVariableSize) : mVariableModuleGlobal->MaxVariableSize);

(2) In the same file, function VariableServiceSetVariable(), we have the
following size checks:

>   //
>   //  The size of the VariableName, including the Unicode Null in bytes plus
>   //  the DataSize is limited to maximum size of PcdGet32 (PcdMaxHardwareErrorVariableSize)
>   //  bytes for HwErrRec#### variable.
>   //
>   if ((Attributes & EFI_VARIABLE_HARDWARE_ERROR_RECORD) == EFI_VARIABLE_HARDWARE_ERROR_RECORD) {
>     if (StrSize (VariableName) + PayloadSize > PcdGet32 (PcdMaxHardwareErrorVariableSize) - GetVariableHeaderSize ()) {
>       return EFI_INVALID_PARAMETER;
>     }
>   } else {
>     //
>     //  The size of the VariableName, including the Unicode Null in bytes plus
>     //  the DataSize is limited to maximum size of Max(Auth)VariableSize bytes.
>     //
>     if ((Attributes & VARIABLE_ATTRIBUTE_AT_AW) != 0) {
>       if (StrSize (VariableName) + PayloadSize > mVariableModuleGlobal->MaxAuthVariableSize - GetVariableHeaderSize ()) {
>         return EFI_INVALID_PARAMETER;
>       }
>     } else {
>       if (StrSize (VariableName) + PayloadSize > mVariableModuleGlobal->MaxVariableSize - GetVariableHeaderSize ()) {
>         return EFI_INVALID_PARAMETER;
>       }
>     }
>   }

That is, the PCDs limit the sum of:
- the variable name (including the terminating L'\0'),
- the variable contents,
- the internal header (which carries the attributes as well). In our
  case that's AUTHENTICATED_VARIABLE_HEADER, because we always format
  the auth varstore GUID into the template.

Furthermore, the 32KB data size for the contents is not based on
"reverse engineering"; the Secure Boot Logo Test tells you exactly what
it is trying to do. (In fact, there is no other way to fix failures in
any of the tests; you can only rely on the messages in order to see what
the test attempted.)

In this case, we have (see
<https://bugzilla.redhat.com/show_bug.cgi?id=1443351#c27>):

> NOTE: if this testcase fails, try restarting the machine and then
> running this test case on its own, for example: "te.exe
> UefiSecureBootLogo.dll
> /name:Microsoft.UefiSecureBootLogo.Tests.ConfirmSetOfLargeVariable"
>
> Test creation of 1 large variable of size 32768 bytes.
>
> Test creation of testLargeVariable Unexpected Error - SetVariable
> failed with status 0xC000000D VariableName "testLargeVariable"
> VendorGuid 77fa9abd-0359-4d32-bd60-28f4e78f784b Attributes 0x00000007
> DataSize 0x00008000

It reports all the parameters of the gRT->SetVariable() call that
failed, except "Data".

In the successful test (with this patch applied), the messages are (see
<https://bugzilla.redhat.com/attachment.cgi?id=1275172>):

> NOTE: if this testcase fails, try restarting the machine and then
> running this test case on its own, for example: "te.exe
> UefiSecureBootLogo.dll
> /name:Microsoft.UefiSecureBootLogo.Tests.ConfirmSetOfLargeVariable"
>
> Test creation of 1 large variable of size 32768 bytes.
>
> AreEqual(7, 7) - testLargeVariable: Attributes from GetVariable()
> match SetVariable()
>
> AreEqual(32768, 32768) - testLargeVariable: DataSize from
> GetVariable() matches SetVariable()

Thanks
Laszlo

>
>> installed both the Windows Server 2008 R2 SP1 test controller and the
>> Windows 2016 Server test client just the other day, together with the
>> most recent filters, using the following descriptions:
>>
>> https://msdn.microsoft.com/en-us/library/windows/hardware/jj123537.aspx
>> https://github.com/daynix/VirtHCK/wiki#Checklist_for_a_New_Controller_VM
>> https://github.com/daynix/VirtHCK/wiki#Checklist_for_a_New_Client_VM
>>
>> Given that this limit can be bumped without breaking compatibility, as
>> you say, I'd like to remain frugal with it, same as we were in James's
>> commit f5404a3eba1d ("OvmfPkg: Increase the maximum size for
>> Authenticated variables", 2016-03-24).
>>
>> I don't understand why the plugfest presentation and the SB Logo Test
>> require different limits... But, I'm certain our QE will find out in
>> short order once the SB Logo Test catches up with the presentation, and
>> I expect I'll submit the corresponding patch soon after.
>>
>> I dislike the speculation in this series, but breaking compatibility is
>> even worse. (A lot worse, to me at least.) So I consider the varstore
>> restructuring the smaller of two wrongs. However, wrt.
>> PcdMaxVariableSize, it seems we're not being forced to either of those
>> wrongs (i.e., breaking compat or speculation), so we can delay the increase.
>>
>>>
>>> Aside from this question:
>>>
>>> Series Reviewed-by: Jordan Justen <jordan.l.justen@intel.com>
>>
>> Thanks a lot!
>>
>> I'll await your ACK for the above argument before pushing the series.
>>
>> Thanks,
>> Laszlo
>>
>>>> In the FD_SIZE_4MB build, our live varstore is now 256 KB big, so we can
>>>> accommodate this. Set both PcdMaxVariableSize and PcdMaxAuthVariableSize
>>>> to 0x8400 -- beyond DataSize=0x8000 from the HCK test, we need some room
>>>> for the variable name and attributes as well.
>>>>
>>>> Cc: Gary Ching-Pang Lin <glin@suse.com>
>>>> Cc: Jordan Justen <jordan.l.justen@intel.com>
>>>> Contributed-under: TianoCore Contribution Agreement 1.0
>>>> Signed-off-by: Laszlo Ersek <lersek@redhat.com>
>>>> ---
>>>>
>>>> Notes:
>>>>     v2:
>>>>     - adjust to FD_SIZE_IN_KB
>>>>     - update commit msg to state 256 KB for the varstore [Jordan]
>>>>
>>>>  OvmfPkg/OvmfPkgIa32.dsc    | 6 ++++++
>>>>  OvmfPkg/OvmfPkgIa32X64.dsc | 6 ++++++
>>>>  OvmfPkg/OvmfPkgX64.dsc     | 6 ++++++
>>>>  3 files changed, 18 insertions(+)
>>>>
>>>> diff --git a/OvmfPkg/OvmfPkgIa32.dsc b/OvmfPkg/OvmfPkgIa32.dsc
>>>> index 26b807dde9fa..e0779ddaa426 100644
>>>> --- a/OvmfPkg/OvmfPkgIa32.dsc
>>>> +++ b/OvmfPkg/OvmfPkgIa32.dsc
>>>> @@ -404,28 +404,34 @@ [PcdsFeatureFlag]
>>>>    gUefiOvmfPkgTokenSpaceGuid.PcdSecureBootEnable|TRUE
>>>>  !endif
>>>>  !if $(SMM_REQUIRE) == TRUE
>>>>    gUefiOvmfPkgTokenSpaceGuid.PcdSmmSmramRequire|TRUE
>>>>    gUefiCpuPkgTokenSpaceGuid.PcdCpuSmmEnableBspElection|FALSE
>>>>  !endif
>>>>
>>>>  [PcdsFixedAtBuild]
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdStatusCodeMemorySize|1
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdResetOnMemoryTypeInformationChange|FALSE
>>>>    gEfiMdePkgTokenSpaceGuid.PcdMaximumGuidedExtractHandler|0x10
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdPeiCoreMaxFvSupported|6
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdPeiCoreMaxPeimPerFv|32
>>>> +!if ($(FD_SIZE_IN_KB) == 1024) || ($(FD_SIZE_IN_KB) == 2048)
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdMaxVariableSize|0x2000
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdMaxAuthVariableSize|0x2800
>>>> +!endif
>>>> +!if $(FD_SIZE_IN_KB) == 4096
>>>> +  gEfiMdeModulePkgTokenSpaceGuid.PcdMaxVariableSize|0x8400
>>>> +  gEfiMdeModulePkgTokenSpaceGuid.PcdMaxAuthVariableSize|0x8400
>>>> +!endif
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdVariableStoreSize|0xe000
>>>>
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdVpdBaseAddress|0x0
>>>>
>>>>    gEfiMdePkgTokenSpaceGuid.PcdReportStatusCodePropertyMask|0x07
>>>>
>>>>    # DEBUG_INIT      0x00000001  // Initialization
>>>>    # DEBUG_WARN      0x00000002  // Warnings
>>>>    # DEBUG_LOAD      0x00000004  // Load events
>>>>    # DEBUG_FS        0x00000008  // EFI File system
>>>>    # DEBUG_POOL      0x00000010  // Alloc & Free (pool)
>>>>    # DEBUG_PAGE      0x00000020  // Alloc & Free (page)
>>>>    # DEBUG_INFO      0x00000040  // Informational debug messages
>>>> diff --git a/OvmfPkg/OvmfPkgIa32X64.dsc b/OvmfPkg/OvmfPkgIa32X64.dsc
>>>> index 41f06a6b6a66..bbe26e2cf452 100644
>>>> --- a/OvmfPkg/OvmfPkgIa32X64.dsc
>>>> +++ b/OvmfPkg/OvmfPkgIa32X64.dsc
>>>> @@ -409,28 +409,34 @@ [PcdsFeatureFlag]
>>>>    gUefiOvmfPkgTokenSpaceGuid.PcdSecureBootEnable|TRUE
>>>>  !endif
>>>>  !if $(SMM_REQUIRE) == TRUE
>>>>    gUefiOvmfPkgTokenSpaceGuid.PcdSmmSmramRequire|TRUE
>>>>    gUefiCpuPkgTokenSpaceGuid.PcdCpuSmmEnableBspElection|FALSE
>>>>  !endif
>>>>
>>>>  [PcdsFixedAtBuild]
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdStatusCodeMemorySize|1
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdResetOnMemoryTypeInformationChange|FALSE
>>>>    gEfiMdePkgTokenSpaceGuid.PcdMaximumGuidedExtractHandler|0x10
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdPeiCoreMaxFvSupported|6
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdPeiCoreMaxPeimPerFv|32
>>>> +!if ($(FD_SIZE_IN_KB) == 1024) || ($(FD_SIZE_IN_KB) == 2048)
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdMaxVariableSize|0x2000
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdMaxAuthVariableSize|0x2800
>>>> +!endif
>>>> +!if $(FD_SIZE_IN_KB) == 4096
>>>> +  gEfiMdeModulePkgTokenSpaceGuid.PcdMaxVariableSize|0x8400
>>>> +  gEfiMdeModulePkgTokenSpaceGuid.PcdMaxAuthVariableSize|0x8400
>>>> +!endif
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdVariableStoreSize|0xe000
>>>>
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdVpdBaseAddress|0x0
>>>>
>>>>    gEfiMdePkgTokenSpaceGuid.PcdReportStatusCodePropertyMask|0x07
>>>>
>>>>    # DEBUG_INIT      0x00000001  // Initialization
>>>>    # DEBUG_WARN      0x00000002  // Warnings
>>>>    # DEBUG_LOAD      0x00000004  // Load events
>>>>    # DEBUG_FS        0x00000008  // EFI File system
>>>>    # DEBUG_POOL      0x00000010  // Alloc & Free (pool)
>>>>    # DEBUG_PAGE      0x00000020  // Alloc & Free (page)
>>>>    # DEBUG_INFO      0x00000040  // Informational debug messages
>>>> diff --git a/OvmfPkg/OvmfPkgX64.dsc b/OvmfPkg/OvmfPkgX64.dsc
>>>> index 053c84b685c5..ff795815f65f 100644
>>>> --- a/OvmfPkg/OvmfPkgX64.dsc
>>>> +++ b/OvmfPkg/OvmfPkgX64.dsc
>>>> @@ -409,28 +409,34 @@ [PcdsFeatureFlag]
>>>>    gUefiOvmfPkgTokenSpaceGuid.PcdSecureBootEnable|TRUE
>>>>  !endif
>>>>  !if $(SMM_REQUIRE) == TRUE
>>>>    gUefiOvmfPkgTokenSpaceGuid.PcdSmmSmramRequire|TRUE
>>>>    gUefiCpuPkgTokenSpaceGuid.PcdCpuSmmEnableBspElection|FALSE
>>>>  !endif
>>>>
>>>>  [PcdsFixedAtBuild]
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdStatusCodeMemorySize|1
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdResetOnMemoryTypeInformationChange|FALSE
>>>>    gEfiMdePkgTokenSpaceGuid.PcdMaximumGuidedExtractHandler|0x10
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdPeiCoreMaxFvSupported|6
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdPeiCoreMaxPeimPerFv|32
>>>> +!if ($(FD_SIZE_IN_KB) == 1024) || ($(FD_SIZE_IN_KB) == 2048)
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdMaxVariableSize|0x2000
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdMaxAuthVariableSize|0x2800
>>>> +!endif
>>>> +!if $(FD_SIZE_IN_KB) == 4096
>>>> +  gEfiMdeModulePkgTokenSpaceGuid.PcdMaxVariableSize|0x8400
>>>> +  gEfiMdeModulePkgTokenSpaceGuid.PcdMaxAuthVariableSize|0x8400
>>>> +!endif
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdVariableStoreSize|0xe000
>>>>
>>>>    gEfiMdeModulePkgTokenSpaceGuid.PcdVpdBaseAddress|0x0
>>>>
>>>>    gEfiMdePkgTokenSpaceGuid.PcdReportStatusCodePropertyMask|0x07
>>>>
>>>>    # DEBUG_INIT      0x00000001  // Initialization
>>>>    # DEBUG_WARN      0x00000002  // Warnings
>>>>    # DEBUG_LOAD      0x00000004  // Load events
>>>>    # DEBUG_FS        0x00000008  // EFI File system
>>>>    # DEBUG_POOL      0x00000010  // Alloc & Free (pool)
>>>>    # DEBUG_PAGE      0x00000020  // Alloc & Free (page)
>>>>    # DEBUG_INFO      0x00000040  // Informational debug messages
>>>> --
>>>> 2.9.3
>>>>
>>>>
>>



  reply	other threads:[~2017-05-04 19:27 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-05-03 21:39 [PATCH v2 0/5] OvmfPkg: add FD_SIZE_4MB for Windows HCK SB tests, and for future proofing Laszlo Ersek
2017-05-03 21:39 ` [PATCH v2 1/5] OvmfPkg: introduce the FD_SIZE_IN_KB macro / build flag Laszlo Ersek
2017-05-03 21:39 ` [PATCH v2 2/5] OvmfPkg/OvmfPkg.fdf.inc: extract VARS_LIVE_SIZE and VARS_SPARE_SIZE macros Laszlo Ersek
2017-05-03 21:39 ` [PATCH v2 3/5] OvmfPkg: introduce 4MB flash image (mainly) for Windows HCK Laszlo Ersek
2017-05-05  4:07   ` Laszlo Ersek
2017-05-05  8:57     ` Jordan Justen
2017-05-05 12:07       ` Laszlo Ersek
2017-05-05 15:43         ` Jordan Justen
2017-05-05 16:46           ` Laszlo Ersek
2017-05-03 21:39 ` [PATCH v2 4/5] OvmfPkg: raise max variable size (auth & non-auth) to 33KB for FD_SIZE_4MB Laszlo Ersek
2017-05-04 16:36   ` Jordan Justen
2017-05-04 16:52     ` Laszlo Ersek
2017-05-04 18:50       ` Jordan Justen
2017-05-04 19:27         ` Laszlo Ersek [this message]
2017-05-04 19:30           ` Laszlo Ersek
2017-05-04 23:00       ` Laszlo Ersek
2017-05-03 21:39 ` [PATCH v2 5/5] OvmfPkg: make the 4MB flash size the default 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=c6f4983e-be93-93ac-4d0e-e1ce42e43f66@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