public inbox for devel@edk2.groups.io
 help / color / mirror / Atom feed
From: Laszlo Ersek <lersek@redhat.com>
To: "Gao, Liming" <liming.gao@intel.com>,
	edk2-devel-01 <edk2-devel@lists.01.org>
Cc: "Wu, Jiaxin" <jiaxin.wu@intel.com>,
	"Kinney, Michael D" <michael.d.kinney@intel.com>,
	"Fu, Siyuan" <siyuan.fu@intel.com>
Subject: Re: [PATCH 02/13] MdePkg/Include/Protocol/Tls.h: pack structures from the TLS RFC
Date: Wed, 4 Apr 2018 12:32:35 +0200	[thread overview]
Message-ID: <e64d7e51-acc3-53dd-51de-83b6de1199fe@redhat.com> (raw)
In-Reply-To: <4A89E2EF3DFEDB4C8BFDE51014F606A14E1F1B10@SHSMSX104.ccr.corp.intel.com>

On 04/03/18 17:08, Gao, Liming wrote:
> Laszlo:
>   Could you use one pack to scope all structure definitions?
I didn't do that originally because the affected structure definitions
are spread out over a larger part of the code, intermixed with enums,
#defines, and other structures, *and* it is not clear from the source
code comments whether *all* of the structure definitions come from RFC
5246. Wherever the code comments did *not* indicate that a given
structure came from the RFC, I didn't add packing either.

Example structures for this:
- EFI_TLS_VERSION
- EFI_TLS_FRAGMENT_DATA

Looking at these in more detail, I don't think either of them should be
packed:

- EfiTlsVersion / EFI_TLS_VERSION are used in TlsSetSessionData()
[NetworkPkg/TlsDxe/TlsProtocol.c], and the constituent fields Major and
Minor are passed individually to TlsSetVersion()
[CryptoPkg/Library/TlsLib/TlsConfig.c]. So packing the structure is not
necessary.

- EFI_TLS_FRAGMENT_DATA contains a (VOID*) pointer, so it cannot be
serialized to the wire by definition. And, it is used in
TlsProcessPacket() [NetworkPkg/TlsDxe/TlsProtocol.c], and
TlsEncryptPacket() / TlsDecryptPacket() [NetworkPkg/TlsDxe/TlsImpl.c],
without any need for packing.

Given that this protocol header file defines both packed and non-packed
structures, without a very clear separation between them, I feel it is
cleaner to use the pack #pragma for individual structures.

If you disagree, then I think we should implement that "clear
separation" in the header file first. Namely, move all the packed
structures either after or before all the non-packed structures, and add
a very visible comment that "all of the following structures come from
RFC 5246 and need packing". Then we can use a single pack #pragma to
cover them all.

Thanks,
Laszlo

>> -----Original Message-----
>> From: Laszlo Ersek [mailto:lersek@redhat.com]
>> Sent: Tuesday, April 3, 2018 10:52 PM
>> To: edk2-devel-01 <edk2-devel@lists.01.org>
>> Cc: Wu, Jiaxin <jiaxin.wu@intel.com>; Gao, Liming <liming.gao@intel.com>; Kinney, Michael D <michael.d.kinney@intel.com>; Fu,
>> Siyuan <siyuan.fu@intel.com>
>> Subject: [PATCH 02/13] MdePkg/Include/Protocol/Tls.h: pack structures from the TLS RFC
>>
>> The structures defined in RFC 5246 are not to have any padding between
>> fields or at the end; use the "pack" pragma as necessary.
>>
>> Cc: Jiaxin Wu <jiaxin.wu@intel.com>
>> Cc: Liming Gao <liming.gao@intel.com>
>> Cc: Michael D Kinney <michael.d.kinney@intel.com>
>> Cc: Siyuan Fu <siyuan.fu@intel.com>
>> Ref: https://bugzilla.tianocore.org/show_bug.cgi?id=915
>> Contributed-under: TianoCore Contribution Agreement 1.1
>> Signed-off-by: Laszlo Ersek <lersek@redhat.com>
>> ---
>>  MdePkg/Include/Protocol/Tls.h | 10 ++++++++++
>>  1 file changed, 10 insertions(+)
>>
>> diff --git a/MdePkg/Include/Protocol/Tls.h b/MdePkg/Include/Protocol/Tls.h
>> index 2119f33c0f5b..dafaabcd2a8b 100644
>> --- a/MdePkg/Include/Protocol/Tls.h
>> +++ b/MdePkg/Include/Protocol/Tls.h
>> @@ -138,33 +138,37 @@ typedef enum {
>>  ///
>>  /// EFI_TLS_CIPHER
>>  /// Note: The definition of EFI_TLS_CIPHER definition is from "RFC 5246, A.4.1.
>>  ///       Hello Messages". The value of EFI_TLS_CIPHER is from TLS Cipher
>>  ///       Suite Registry of IANA.
>>  ///
>> +#pragma pack (1)
>>  typedef struct {
>>    UINT8                         Data1;
>>    UINT8                         Data2;
>>  } EFI_TLS_CIPHER;
>> +#pragma pack ()
>>
>>  ///
>>  /// EFI_TLS_COMPRESSION
>>  /// Note: The value of EFI_TLS_COMPRESSION definition is from "RFC 3749".
>>  ///
>>  typedef UINT8 EFI_TLS_COMPRESSION;
>>
>>  ///
>>  /// EFI_TLS_EXTENSION
>>  /// Note: The definition of EFI_TLS_EXTENSION if from "RFC 5246 A.4.1.
>>  ///       Hello Messages".
>>  ///
>> +#pragma pack (1)
>>  typedef struct {
>>    UINT16                        ExtensionType;
>>    UINT16                        Length;
>>    UINT8                         Data[1];
>>  } EFI_TLS_EXTENSION;
>> +#pragma pack ()
>>
>>  ///
>>  /// EFI_TLS_VERIFY
>>  /// Use either EFI_TLS_VERIFY_NONE or EFI_TLS_VERIFY_PEER, the last two options
>>  /// are 'ORed' with EFI_TLS_VERIFY_PEER if they are desired.
>>  ///
>> @@ -191,35 +195,41 @@ typedef UINT32  EFI_TLS_VERIFY;
>>
>>  ///
>>  /// EFI_TLS_RANDOM
>>  /// Note: The definition of EFI_TLS_RANDOM is from "RFC 5246 A.4.1.
>>  ///       Hello Messages".
>>  ///
>> +#pragma pack (1)
>>  typedef struct {
>>    UINT32                        GmtUnixTime;
>>    UINT8                         RandomBytes[28];
>>  } EFI_TLS_RANDOM;
>> +#pragma pack ()
>>
>>  ///
>>  /// EFI_TLS_MASTER_SECRET
>>  /// Note: The definition of EFI_TLS_MASTER_SECRET is from "RFC 5246 8.1.
>>  ///       Computing the Master Secret".
>>  ///
>> +#pragma pack (1)
>>  typedef struct {
>>    UINT8                         Data[48];
>>  } EFI_TLS_MASTER_SECRET;
>> +#pragma pack ()
>>
>>  ///
>>  /// EFI_TLS_SESSION_ID
>>  /// Note: The definition of EFI_TLS_SESSION_ID is from "RFC 5246 A.4.1. Hello Messages".
>>  ///
>>  #define MAX_TLS_SESSION_ID_LENGTH  32
>> +#pragma pack (1)
>>  typedef struct {
>>    UINT16                        Length;
>>    UINT8                         Data[MAX_TLS_SESSION_ID_LENGTH];
>>  } EFI_TLS_SESSION_ID;
>> +#pragma pack ()
>>
>>  ///
>>  /// EFI_TLS_SESSION_STATE
>>  ///
>>  typedef enum {
>>    ///
>> --
>> 2.14.1.3.gb7cf6e02401b
>>
> 



  reply	other threads:[~2018-04-04 10:32 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-04-03 14:51 [PATCH 00/13] {Ovmf, Mde, Network, Crypto}Pkg: fixes+features for setting HTTPS cipher suites Laszlo Ersek
2018-04-03 14:51 ` [PATCH 01/13] OvmfPkg/TlsAuthConfigLib: configure trusted cipher suites for HTTPS boot Laszlo Ersek
2018-04-03 14:51 ` [PATCH 02/13] MdePkg/Include/Protocol/Tls.h: pack structures from the TLS RFC Laszlo Ersek
2018-04-03 15:08   ` Gao, Liming
2018-04-04 10:32     ` Laszlo Ersek [this message]
2018-04-10  1:51   ` Fu, Siyuan
2018-04-03 14:51 ` [PATCH 03/13] NetworkPkg/TlsDxe: verify DataSize for EfiTlsCipherList Laszlo Ersek
2018-04-10  1:51   ` Fu, Siyuan
2018-04-03 14:51 ` [PATCH 04/13] NetworkPkg/TlsDxe: clean up byte order conversion " Laszlo Ersek
2018-04-10  1:53   ` Fu, Siyuan
2018-04-03 14:51 ` [PATCH 05/13] CryptoPkg/TlsLib: replace TlsGetCipherString() with TlsGetCipherMapping() Laszlo Ersek
2018-04-03 14:51 ` [PATCH 06/13] CryptoPkg/TlsLib: use binary search in the TlsGetCipherMapping() function Laszlo Ersek
2018-04-03 14:51 ` [PATCH 07/13] CryptoPkg/TlsLib: pre-compute OpensslCipherLength in TlsCipherMappingTable Laszlo Ersek
2018-04-03 14:51 ` [PATCH 08/13] CryptoPkg/TlsLib: add the "TlsMappingTable.sh" POSIX shell script Laszlo Ersek
2018-04-03 14:51 ` [PATCH 09/13] CryptoPkg/TlsLib: extend "TlsCipherMappingTable" Laszlo Ersek
2018-04-03 14:51 ` [PATCH 10/13] CryptoPkg/TlsLib: sort [LibraryClasses] section in the INF file Laszlo Ersek
2018-04-03 14:51 ` [PATCH 11/13] CryptoPkg/TlsLib: sanitize lib classes in internal header and INF Laszlo Ersek
2018-04-03 14:51 ` [PATCH 12/13] CryptoPkg/TlsLib: clean up leading comment for TlsSetCipherList() Laszlo Ersek
2018-04-03 14:51 ` [PATCH 13/13] CryptoPkg/TlsLib: rewrite TlsSetCipherList() Laszlo Ersek
2018-04-10  4:09 ` [PATCH 00/13] {Ovmf, Mde, Network, Crypto}Pkg: fixes+features for setting HTTPS cipher suites Wu, Jiaxin
2018-04-10  7:40   ` Long, Qin
2018-04-10 10:02     ` Laszlo Ersek
2018-04-10 10:10       ` Laszlo Ersek
2018-04-10 16:56         ` Long, Qin
2018-04-10  9:47   ` Laszlo Ersek
2018-04-10 17:06     ` Long, Qin
2018-04-10 20:06       ` Laszlo Ersek
2018-04-11  1:59         ` Wu, Jiaxin

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=e64d7e51-acc3-53dd-51de-83b6de1199fe@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