From: "Rich" 

This is a multi-part message in MIME format.

------=_NextPart_000_0014_01C77B81.A0376110
Content-Type: text/plain;
        charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

   Have these folks never spoken with an engineer?  Of course this was =
tested and yes if emitted everywhere it is unacceptable.  In performance =
for sure and I think size too but I don't remember.

   As for a paranoid option as suggested, this is available via =
different means using #pragma strict_gs_check which Michael Howard = describes at =
http://blogs.msdn.com/michael_howard/archive/2007/04/03/hardening-stack-b=
ased-buffer-overrun-detection-in-vc-2005-sp1.aspx.  It's not as = described
as emitting one with no need.  It uses a different and lower = bar for when
one should be emitted.

   Both with and without the pragma checks are emitted when one is not =
needed and both could miss unusual cases where one could be needed = though
that is less likely with the pragma because it is a more = inclusive test. 
Then again, I think these folks don't understand the = purpose of the
checks.  They are not intended to fix or avoid bugs.  = They are a
mitigation for when bugs do exist to limit the risk due to = the bug.

Rich

  "mike"  wrote in message =
news:e61o13pfa398b42o3shukiet2je1cjoe3b{at}4ax.com...


  http://www.eweek.com/article2/0,1895,2110151,00.asp



  =3D=3D=3D
  Once again, obviously an engineering trade-off decision was made.
  Perhaps Microsoft was concerned that putting in stack checking =
literally
  everywhere would fatten the program up beyond what was acceptable. Its
  compiler, its source code, it could do the testing to see.

   It seems to me that
  maybe there's a need for a /GSP (for "Paranoid") switch that puts in =
the
  stack check even if there doesn't seem to be a need, unless there's a
  good reason not to (there are cases, described by Microsoft, where the
  stack check code wouldn't be reliable).

  =3D=3D=3D


   /m
------=_NextPart_000_0014_01C77B81.A0376110
Content-Type: text/html;
        charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable








   Have
these folks never =
spoken with an=20
engineer?  Of course this was tested and yes if emitted everywhere
= it is=20
unacceptable.  In performance for sure and I think size too but I = don't=20
remember.
 
   As for a
paranoid option =
as suggested,=20
this is available via different means using #pragma strict_gs_check = which=20
Michael Howard describes at http://blogs.msdn.com/michael_howard/archive/2007/04/03/hardening=
-stack-based-buffer-overrun-detection-in-vc-2005-sp1.aspx">http://blogs.m=
sdn.com/michael_howard/archive/2007/04/03/hardening-stack-based-buffer-ov=
errun-detection-in-vc-2005-sp1.aspx. =20
It's not as described as emitting one with no need.  It uses a =
different=20
and lower bar for when one should be emitted.
 
   Both with
and without the =
pragma=20
checks are emitted when one is not needed and both could miss unusual = cases=20
where one could be needed though that is less likely with the pragma = because it=20
is a more inclusive test.  Then again, I think these folks don't =
understand=20
the purpose of the checks.  They are not intended to fix or
avoid=20 bugs.  They are a mitigation for when bugs do exist to
limit the = risk due=20
to the bug.
 
Rich
 

  "mike" <mike{at}barkto.com>">mailto:mike{at}barkto.com">mike{at}barkto.com>
wrote=20
  in message news:e61o13pfa39=
8b42o3shukiet2je1cjoe3b{at}4ax.com...
  http://www.e" target="new">http://www.e=">http://www.eweek.com/article2/0,1895,2110151,00.asp">http://www.e=
week.com/article2/0,1895,2110151,00.asp=3D=3D=3DO=
nce=20
  again, obviously an engineering trade-off decision was =
made.Perhaps=20
  Microsoft was concerned that putting in stack checking =
literallyeverywhere=20
  would fatten the program up beyond what was acceptable. =
Itscompiler, its=20
  source code, it could do the testing to see.
   
   It seems to me thatmaybe there's a need
for a /GSP (for =

  "Paranoid") switch that puts in thestack check even if there =
doesn't seem=20
  to be a need, unless there's agood reason not to (there are cases, =

  described by Microsoft, where thestack check code wouldn't be=20
  reliable).
 
=3D=3D=3D /m

------=_NextPart_000_0014_01C77B81.A0376110--

--- BBBS/NT v4.01 Flag-5