full-disclosure - Re: Defense in depth -- the Microsoft way (part 13): surprising and inconsistent behaviour, sloppy coding, sloppy QA, sloppy documentation

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: <CAJB2JzsL92mkJnHAhP0wgRyPiMWpcg8YY6cFzrd+8K_yw9g1Yg@mail.gmail.com>
Date: Sun, 3 Nov 2013 17:38:02 -0200
From: Mario Vilas <mvilas@...il.com>
To: Stefan Kanthak <stefan.kanthak@...go.de>
Cc: "full-disclosure@...ts.grok.org.uk" <full-disclosure@...ts.grok.org.uk>,
 bugtraq <bugtraq@...urityfocus.com>
Subject: Re: Defense in depth -- the Microsoft way (part
 13): surprising and inconsistent behaviour, sloppy coding, sloppy QA,
 sloppy documentation

This may be a silly question, so I apologize in advance, but that would
exactly be the advantage here? Using a NULL pointer is in most (if not all)
those cases undocumented behavior to begin with. Unless I'm missing
something, the problem is not so much with Win32 as it is with the C
language in general...


On Sun, Nov 3, 2013 at 4:30 PM, Stefan Kanthak <stefan.kanthak@...go.de>wrote:

> Hi @ll,
>
> the Win32 API is full of idiosyncrasies resp. surprising and inconsistent,
> poorly tested and documented behaviour.
>
> Just to pick one: NULL pointer as string argument.
>
> 0. lstrlen(NULL)
>    lstrcat(NULL, ...)      and  lstrcat(..., NULL)
>    lstrcmp(NULL, ...)      and  lstrcmp(..., NULL)
>    lstrcmpi(NULL, ...)     and  lstrcmpi(..., NULL)
>    lstrcpy(NULL, ...)      and  lstrcpy(..., NULL)
>    lstrcpyn(NULL, ..., 0)  and  lstrcpy(..., NULL, ...)
>
>    do not yield an exception, but treat their NULL arguments like an
>    empty string (when used as source), resp. return NULL (when used as
>    destination).
>
>
> 1. wsprintf(NULL, ...)       and  wvsprintf(NULL, ...)
>    wsprintf(..., NULL, ...)  and  wvsprintf(..., NULL, ...)
>
>    yield an access violation in USER32.DLL.
>
>
> 2. CommandLineToArgvW(NULL, ...)
>
>    yields an access violation in SHELL32.DLL.
>
>
> 3. CreateProcess(NULL, NULL, ...)
>    CreateProcessAsUser(..., NULL, NULL, ...)
>    CreateProcessWithLogonW(..., ..., ..., ..., NULL, NULL, ...)
>    CreateProcessWithTokenW(..., ..., NULL, NULL, ...)
>
>    yield an access violation in KERNEL32.DLL.
>
>
> 4. GetFileAttributes(NULL)
>
>    does not yield an exception, but treats the NULL argument like an
>    empty string.
>
>
> 5. GetBinaryType(NULL, ...)
>
>    does not yield an exception, but treats the NULL argument like an
>    empty string.
>
>
> 6. MessageBox(..., NULL, ...)  and  MessageBox(..., ..., NULL, ...)
>
>    do not yield an exception, but treat the NULL argument like an
>    empty string.
>
>
> 7. FatalAppExit(0, NULL)
>
>    does not yield an exception, but treats the NULL argument like an
>    empty string.
>
>
> 8. GetCurrentDirectory(..., NULL)
>
>    returns an error if the buffer size (the argument shown as ... here)
>    is sufficient to hold the result, else the required buffer size.
>
>    GetTempPath(..., NULL)
>    GetSystemDirectory(NULL, ...)
>    GetSystemWindowsDirectory(NULL, ...)
>    GetSystemWow64Directory(NULL, ...)
>    GetWindowsDirectory(NULL, ...)
>    GetComputerName(NULL, ...)
>
>    yield an access violation in NTDLL.DLL resp. KERNEL32.DLL if the
>    buffer size is sufficient to hold the result, else the required
>    buffer size.
>
>    GetUserName(NULL, ...)
>    GetComputerObjectName(..., NULL, ...)
>
>    do not yield an access violation, but return an error with
>    GetLastError() == ERROR_INSUFFICIENT_BUFFER.
>
>
> 9. GetUserName(NULL, NULL)
>    GetComputerName(NULL, NULL)
>
>    yield an access violation in KERNEL32.DLL.
>
>    GetComputerNameEx(..., NULL, NULL)
>    GetComputerObjectName(..., NULL, NULL)
>
>    do not yield an access violation, but return an error with
>    GetLastError() == ERROR_INVALID_PARAMETER.
>
>    JFTR: only the documentation of the last function (see
>          <http://msdn.microsoft.com/en-us/library/ms724301.aspx>)
>          explicitly says about the value of the third argument
>          "If lpBuffer is NULL, this parameter must be zero."
>          and checks this contraint properly.
>
>
> The expected behavior in all cases is but to return an error with
> GetLastError() == ERROR_INVALID_PARAMETER or similar.
>
>
> FIX: ALL interfaces of the Win32 API should^WMUST verify (ALL) their
>      arguments properly before using them and return an appropriate,
>      documented error code.
>
>
> stay tuned
> Stefan Kanthak
>
> _______________________________________________
> Full-Disclosure - We believe in it.
> Charter: http://lists.grok.org.uk/full-disclosure-charter.html
> Hosted and sponsored by Secunia - http://secunia.com/
>



-- 
“There's a reason we separate military and the police: one fights the enemy
of the state, the other serves and protects the people. When the military
becomes both, then the enemies of the state tend to become the people.”

Content of type "text/html" skipped

_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/