From: "Vitaly Cheptsov" <vit9696@protonmail.com>
To: "Yao, Jiewen" <jiewen.yao@intel.com>
Cc: "devel@edk2.groups.io" <devel@edk2.groups.io>,
"Gao, Liming" <liming.gao@intel.com>,
"Wang, Jian J" <jian.j.wang@intel.com>,
"Kinney, Michael D" <michael.d.kinney@intel.com>,
Laszlo Ersek <lersek@redhat.com>,
Marvin.Haeuser@outlook.com
Subject: Re: [edk2-devel] [PATCH v1 1/1] MdePkg: Add PCD to disable safe string constraint assertions
Date: Mon, 21 Oct 2019 08:29:50 +0000 [thread overview]
Message-ID: <CDA85E7C-AE0B-4BEC-83B0-2957CC90FB6B@protonmail.com> (raw)
In-Reply-To: <74D8A39837DF1E4DA445A8C0B3885C503F80B961@shsmsx102.ccr.corp.intel.com>
[-- Attachment #1: Type: text/plain, Size: 12217 bytes --]
Jiewen,
Your explanation makes good sense in the context of "This API is NOT design to handle untrusted input.", however, I believe this is not how it is should work or at least this is not how we would like it to behave.
In Unix world there are similar hardened interfaces, for example, strlcpy or strlcat[1]. These interfaces behave quite similar to what you have in BaseLib SafeString, but in the first place they are meant to handle untrusted input. To my experience this situation happens no less often (and in fact much more) than the need to do extra error checking in trusted code, and I expect EDK2 to follow suit. I.e. implement something that can be used not only to check for programmer errors within the module, but also perform the validation of external data. After all, if these were just the assertions, I would have expected for return value to be void not RETURN_STATUS.
If we consider all the options we need to do either of the following:
- reimplement most of the functions in the caller code, which is non-trivial, increases code size, reduces readability, is more prone to errors, makes reviewing more time consuming, etc.
- reimplement all these functions in a separate library, which solves some of the issues, but still increases code size and results in separate interfaces with different contracts
- add a pcd to disable assertions, which will disable the debugging handlers and will effectively make these functions behave as runtime checks for untrusted data, numerous people me included expect them do.
Last suggestion makes most sense to me. Therefore, I believe that even in the situation we have different opinions on whether this should be on by default, we should at least see the benefit for having an option to make this configurable in other projects. In this case I can update the patch in the next days to preserve the original behaviour and resubmit it as a v2.
Best wishes,
Vitaly
[1] https://linux.die.net/man/3/strlcpy
> 21 окт. 2019 г., в 11:07, Yao, Jiewen <jiewen.yao@intel.com> написал(а):
>
>
> Hi Vitaly
> We have discussed the ASSERT usage when we added the first version of code.
>
> C11 standard supports runtime violation handler registration.
> We investigated the behavior of
> (1) MS Visual Studio - http://msdn.microsoft.com/en-us/library/bb288454.aspx
> (2) SafeCLib - http://sourceforge.net/projects/safeclib/
> (3) Slibc - https://code.google.com/p/slibc/
>
> The default behavior is:
> (1) Call Debugger
> (2) Print error
> (3) Call abort()
>
> As conclusion, we believe it is *caller's responsibility* to make sure the caller inputs right parameter, instead of let callee check and return error.
>
> In order to catch such error as early as possible, we decide to use ASSERT, because it is something never happen. (We still use error handling followed by to handle the release build.)
>
> This API is NOT design to handle untrusted input. As such, we believe it is expected ASSERT behavior.
>
>
> We have other debate, such as if we need ASSERT 2 bytes alignment CHAR16 process, or if we need ASSERT address alignment for MMIO access.
> Per our experience, it is much better to let caller guarantee that, instead of callee check that. And ASSERT is a good way to catch the issue as early as possible.
>
>
>
> Thank you
> Yao Jiewen
>
>> -----Original Message-----
>> From: devel@edk2.groups.io <devel@edk2.groups.io> On Behalf Of Vitaly
>> Cheptsov via Groups.Io
>> Sent: Monday, October 21, 2019 3:28 PM
>> To: Yao, Jiewen <jiewen.yao@intel.com>; Gao, Liming <liming.gao@intel.com>
>> Cc: devel@edk2.groups.io; Wang, Jian J <jian.j.wang@intel.com>; Kinney,
>> Michael D <michael.d.kinney@intel.com>
>> Subject: Re: [edk2-devel] [PATCH v1 1/1] MdePkg: Add PCD to disable safe string
>> constraint assertions
>>
>> Liming, Jiewen,
>>
>> I am personally fine to resubmit the patch changing the defaults to TRUE, but
>> does it actually make sense in any other environment but some special testing
>> platform? I cannot imagine any production platform that would need it enabled,
>> the only use case is to perform analysis on the trusted data usage or something
>> similar, as I explained on BZ.
>>
>> As for assertions, I would expect that all PCDs we introduce are meant to
>> control unexpected ASSERT behaviour. I.e. where ASSERTs are potentially used
>> to signalise about issues coming not only from trusted sources (usage contract
>> violation) but also from untrusted sources (external data). Such assertions
>> cannot be used in production environments, as they may break software, and
>> thus we have to implement code to disable them.
>>
>> Best regards,
>> Vitaly
>>
>>> 21 окт. 2019 г., в 7:28, Yao, Jiewen <jiewen.yao@intel.com> написал(а):
>>>
>>>
>>> Hi Mike
>>> I remember we have discussed it before.
>>>
>>> The general concern is that: how many additional PCD we need introduce, to
>> control different ASSERT in different modules ?
>>>
>>> We may want to enable *some* assert in some modules, but disable *some
>> other* assert.
>>> E.g. the assert for linkedlist in base lib is another example...
>>>
>>> What is your thought?
>>>
>>> Thank you
>>> Yao Jiewen
>>>
>>>> -----Original Message-----
>>>> From: Gao, Liming <liming.gao@intel.com>
>>>> Sent: Monday, October 21, 2019 11:17 AM
>>>> To: devel@edk2.groups.io; vit9696@protonmail.com
>>>> Cc: Yao, Jiewen <jiewen.yao@intel.com>; Wang, Jian J
>> <jian.j.wang@intel.com>;
>>>> Gao, Liming <liming.gao@intel.com>; Kinney, Michael D
>>>> <michael.d.kinney@intel.com>
>>>> Subject: RE: [edk2-devel] [PATCH v1 1/1] MdePkg: Add PCD to disable safe
>> string
>>>> constraint assertions
>>>>
>>>> Include more people.
>>>>
>>>> Basically, to keep the compatible behavior,
>> PcdAssertOnSafeStringConstraints
>>>> default value should be TRUE.
>>>> The different platform can configure it.
>>>>
>>>> Thanks
>>>> Liming
>>>>> -----Original Message-----
>>>>> From: devel@edk2.groups.io [mailto:devel@edk2.groups.io] On Behalf Of
>>>>> Vitaly Cheptsov via Groups.Io
>>>>> Sent: Sunday, October 20, 2019 9:06 PM
>>>>> To: devel@edk2.groups.io
>>>>> Subject: [edk2-devel] [PATCH v1 1/1] MdePkg: Add PCD to disable safe
>> string
>>>>> constraint assertions
>>>>>
>>>>> REF: https://bugzilla.tianocore.org/show_bug.cgi?id=2054
>>>>>
>>>>> Runtime data checks are not meant to cause debug assertions
>>>>> unless explicitly needed by some debug code (thus the PCD)
>>>>> as this breaks debug builds validating data with BaseLib.
>>>>>
>>>>> Signed-off-by: Vitaly Cheptsov <vit9696@protonmail.com>>
>>>>> ---
>>>>> MdePkg/MdePkg.dec | 6 ++++++
>>>>> MdePkg/Library/BaseLib/BaseLib.inf | 11 ++++++-----
>>>>> MdePkg/Library/BaseLib/SafeString.c | 4 +++-
>>>>> MdePkg/MdePkg.uni | 6 ++++++
>>>>> 4 files changed, 21 insertions(+), 6 deletions(-)
>>>>>
>>>>> diff --git a/MdePkg/MdePkg.dec b/MdePkg/MdePkg.dec
>>>>> index 3fd7d1634c..dda2cdf401 100644
>>>>> --- a/MdePkg/MdePkg.dec
>>>>> +++ b/MdePkg/MdePkg.dec
>>>>> @@ -2221,6 +2221,12 @@ [PcdsFixedAtBuild,PcdsPatchableInModule]
>>>>> # @Prompt Memory Address of GuidedExtractHandler Table.
>>>>>
>>>>>
>> gEfiMdePkgTokenSpaceGuid.PcdGuidedExtractHandlerTableAddress|0x10000
>>>>> 00|UINT64|0x30001015
>>>>>
>>>>> + ## Indicates if safe string constraint violation should assert.<BR><BR>
>>>>> + # TRUE - Safe string constraint violation causes assertion.<BR>
>>>>> + # FALSE - Safe string constraint violation does not cause assertion.<BR>
>>>>> + # @Prompt Enable safe string constraint violation assertions.
>>>>> +
>>>>>
>> gEfiMdePkgTokenSpaceGuid.PcdAssertOnSafeStringConstraints|FALSE|BOOL
>>>>> EAN|0x0000002e
>>>>> +
>>>>> [PcdsFixedAtBuild, PcdsPatchableInModule, PcdsDynamic, PcdsDynamicEx]
>>>>> ## This value is used to set the base address of PCI express hierarchy.
>>>>> # @Prompt PCI Express Base Address.
>>>>> diff --git a/MdePkg/Library/BaseLib/BaseLib.inf
>>>>> b/MdePkg/Library/BaseLib/BaseLib.inf
>>>>> index 3586beb0ab..bc98bc6134 100644
>>>>> --- a/MdePkg/Library/BaseLib/BaseLib.inf
>>>>> +++ b/MdePkg/Library/BaseLib/BaseLib.inf
>>>>> @@ -390,11 +390,12 @@ [LibraryClasses]
>>>>> BaseMemoryLib
>>>>>
>>>>> [Pcd]
>>>>> - gEfiMdePkgTokenSpaceGuid.PcdMaximumLinkedListLength ##
>>>>> SOMETIMES_CONSUMES
>>>>> - gEfiMdePkgTokenSpaceGuid.PcdMaximumAsciiStringLength ##
>>>>> SOMETIMES_CONSUMES
>>>>> - gEfiMdePkgTokenSpaceGuid.PcdMaximumUnicodeStringLength ##
>>>>> SOMETIMES_CONSUMES
>>>>> - gEfiMdePkgTokenSpaceGuid.PcdControlFlowEnforcementPropertyMask
>>>>> ## SOMETIMES_CONSUMES
>>>>> - gEfiMdePkgTokenSpaceGuid.PcdSpeculationBarrierType ##
>>>>> SOMETIMES_CONSUMES
>>>>> + gEfiMdePkgTokenSpaceGuid.PcdAssertOnSafeStringConstraints ##
>>>>> SOMETIMES_CONSUMES
>>>>> + gEfiMdePkgTokenSpaceGuid.PcdMaximumLinkedListLength ##
>>>>> SOMETIMES_CONSUMES
>>>>> + gEfiMdePkgTokenSpaceGuid.PcdMaximumAsciiStringLength ##
>>>>> SOMETIMES_CONSUMES
>>>>> + gEfiMdePkgTokenSpaceGuid.PcdMaximumUnicodeStringLength ##
>>>>> SOMETIMES_CONSUMES
>>>>> + gEfiMdePkgTokenSpaceGuid.PcdControlFlowEnforcementPropertyMask
>>>>> ## SOMETIMES_CONSUMES
>>>>> + gEfiMdePkgTokenSpaceGuid.PcdSpeculationBarrierType ##
>>>>> SOMETIMES_CONSUMES
>>>>>
>>>>> [FeaturePcd]
>>>>> gEfiMdePkgTokenSpaceGuid.PcdVerifyNodeInList ## CONSUMES
>>>>> diff --git a/MdePkg/Library/BaseLib/SafeString.c
>>>>> b/MdePkg/Library/BaseLib/SafeString.c
>>>>> index 7dc03d2caa..56b5e34a8d 100644
>>>>> --- a/MdePkg/Library/BaseLib/SafeString.c
>>>>> +++ b/MdePkg/Library/BaseLib/SafeString.c
>>>>> @@ -14,7 +14,9 @@
>>>>>
>>>>> #define SAFE_STRING_CONSTRAINT_CHECK(Expression, Status) \
>>>>> do { \
>>>>> - ASSERT (Expression); \
>>>>> + if (PcdGetBool (PcdAssertOnSafeStringConstraints)) { \
>>>>> + ASSERT (Expression); \
>>>>> + } \
>>>>> if (!(Expression)) { \
>>>>> return Status; \
>>>>> } \
>>>>> diff --git a/MdePkg/MdePkg.uni b/MdePkg/MdePkg.uni
>>>>> index 5c1fa24065..425b66bb43 100644
>>>>> --- a/MdePkg/MdePkg.uni
>>>>> +++ b/MdePkg/MdePkg.uni
>>>>> @@ -287,6 +287,12 @@
>>>>>
>>>>> #string
>>>>> STR_gEfiMdePkgTokenSpaceGuid_PcdGuidedExtractHandlerTableAddress_H
>>>>> ELP #language en-US "This value is used to set the available memory
>> address
>>>>> to store Guided Extract Handlers. The required memory space is decided by
>>>>> the value of PcdMaximumGuidedExtractHandler."
>>>>>
>>>>> +#string
>>>>> STR_gEfiMdePkgTokenSpaceGuid_PcdAssertOnSafeStringConstraints_PROM
>>>>> PT #language en-US "Enable safe string constraint violation assertions"
>>>>> +
>>>>> +#string
>>>>> STR_gEfiMdePkgTokenSpaceGuid_PcdAssertOnSafeStringConstraints_HELP
>>>>> #language en-US "Indicates if safe string constraint violation should
>>>>> assert.<BR><BR>\n"
>>>>> + "TRUE - Safe string
>> constraint
>>>>> violation causes assertion.<BR>\n"
>>>>> + "FALSE - Safe string
>> constraint
>>>>> violation does not cause assertion.<BR>"
>>>>> +
>>>>> #string
>>>>> STR_gEfiMdePkgTokenSpaceGuid_PcdPciExpressBaseAddress_PROMPT
>>>>> #language en-US "PCI Express Base Address"
>>>>>
>>>>> #string STR_gEfiMdePkgTokenSpaceGuid_PcdPciExpressBaseAddress_HELP
>>>>> #language en-US "This value is used to set the base address of PCI express
>>>>> hierarchy."
>>>>> --
>>>>> 2.21.0 (Apple Git-122)
>>>>>
>>>>>
>>>>> -=-=-=-=-=-=
>>>>> Groups.io Links: You receive all messages sent to this group.
>>>>>
>>>>> View/Reply Online (#49255):
>> https://edk2.groups.io/g/devel/message/49255
>>>>> Mute This Topic: https://groups.io/mt/35943317/1759384
>>>>> Group Owner: devel+owner@edk2.groups.io
>>>>> Unsubscribe: https://edk2.groups.io/g/devel/unsub [liming.gao@intel.com]
>>>>> -=-=-=-=-=-=
>>>
>>
>>
>>
>
[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 489 bytes --]
next prev parent reply other threads:[~2019-10-21 8:30 UTC|newest]
Thread overview: 10+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-10-20 13:05 [PATCH v1 0/1] MdePkg: Add PCD to disable safe string constraint assertions Vitaly Cheptsov
2019-10-20 13:05 ` [PATCH v1 1/1] " Vitaly Cheptsov
2019-10-21 3:16 ` [edk2-devel] " Liming Gao
2019-10-21 4:28 ` Yao, Jiewen
2019-10-21 7:27 ` Vitaly Cheptsov
2019-10-21 8:07 ` Yao, Jiewen
2019-10-21 8:29 ` Vitaly Cheptsov [this message]
2019-10-21 8:51 ` Yao, Jiewen
2019-10-21 9:58 ` Vitaly Cheptsov
2019-10-21 13:46 ` 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=CDA85E7C-AE0B-4BEC-83B0-2957CC90FB6B@protonmail.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