This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Storage for uninitialized objects (PR 24626)
- From: Robert Dewar <dewar at adacore dot com>
- To: John David Anglin <dave at hiauly1 dot hia dot nrc dot ca>
- Cc: Steven Bosscher <steven at gcc dot gnu dot org>, gcc at gcc dot gnu dot org, schlie at comcast dot net, ian at airs dot com, jsm28 at gcc dot gnu dot org
- Date: Sun, 15 Jan 2006 14:43:06 -0500
- Subject: Re: Storage for uninitialized objects (PR 24626)
- References: <200601151935.k0FJZ0OI002472@hiauly1.hia.nrc.ca>
John David Anglin wrote:
This seems pretty clear. C99 requires that storage be allocated
for uninitialized objects, that an indeterminate value be stored
in the object when the declarator for the object is reached in the
block, that the last-stored value be retained for the duration of
the block.
I think that is an incorrect interpretation. Remember the standard is
always "as-if" when it gives an implementation approach. Please
show a correct C program that can tell that gcc is not following
the above scheme.
Now, the argument being advanced here is that this is all irrelevant
since the use of an indeterminate value constitutes undefined behavior
and GCC can pretty much do anything. However, "undefined behavior"
is conditional on the standard not imposing requirements on the construct,
data or object. Further, I think the general intent was that
implementations be as portable as possible.
Obviously the "as possible" is the operable word here, or you would not find
undefined in the standard. Undefined is always about non-portability.
Dave