More on alignment

[H.K. Kingston-Smith]

Thanks to everybody who replied to my question on alignment on
stack arrays. In a similar vein, what can one say about the alignment of
p in

char *p = alloca(N) ;

where N can be any integer smaller than the stack space available? Is it
going to follow similar alignment patterns as malloc(N) or as char p[N]?
I am aware that using alloca() is not encouraged, but I have come across
a few situations in which it would advantageous to do so.

[Kenny McCormack]

In article <g3**********@registered.motzarella.org>,
H.K. Kingston-Smith <HK***@yahoo.comwrote:

> Thanks to everybody who replied to my question on alignment on
st**k arrays. In a similar vein, what can one say about the alignment of
p in

char *p = alloca(N) ;

where N can be any integer smaller than the stack space available? Is it
going to follow similar alignment patterns as malloc(N) or as char p[N]?
I am aware that using alloca() is not encouraged, but I have come across
a few situations in which it would advantageous to do so.

You are now allowed to use that kind of language ("st**k" or "a((oca")
in this newsgroup.

Now, go wash your mouth out with soap.

[Richard Heathfield]

H.K. Kingston-Smith said:

Thanks to everybody who replied to my question on alignment on
stack arrays. In a similar vein, what can one say about the alignment of
p in

char *p = alloca(N) ;

where N can be any integer smaller than the stack space available?

Nothing, really. Since alloca is not standardised, each implementation is free
to implement alloca in whatever way it likes with whatever semantics and
interface it likes, or indeed choose not to implement it at all.

This is one of those times when the only meaningful answer is to consult a
newsgroup that is dedicated to your particular implementation, and don't
assume that the answers you get apply to other implementations.

--
Richard Heathfield <http://www.cpax.org.uk>
Email: -http://www. +rjh@
Google users: <http://www.cpax.org.uk/prg/writings/googly.php>
"Usenet is a strange place" - dmr 29 July 1999

[santosh]

H.K. Kingston-Smith wrote:

Thanks to everybody who replied to my question on alignment on
stack arrays. In a similar vein, what can one say about the alignment
of p in

char *p = alloca(N) ;

where N can be any integer smaller than the stack space available? Is
it going to follow similar alignment patterns as malloc(N) or as char
p[N]? I am aware that using alloca() is not encouraged, but I have
come across a few situations in which it would advantageous to do so.

Strictly speaking alloca is not defined by the C standard (or any other
standard AFAIK) so it's behaviour cannot be accurately determined
without examining your implementation of it. However all the
implementations of alloca that I have seen have returned memory that is
aligned for the strictest possible type in the system. Hence it can
used to initialise any pointer type. In this way it's exactly like
malloc, as far as I can see.

[rahul]

On Jun 16, 10:29 pm, "H.K. Kingston-Smith" <[e-mail]:

Thanks to everybody who replied to my question on alignment on
stack arrays. In a similar vein, what can one say about the alignment of
p in

char *p = alloca(N) ;

where N can be any integer smaller than the stack space available? Is it
going to follow similar alignment patterns as malloc(N) or as char p[N]?
I am aware that using alloca() is not encouraged, but I have come across
a few situations in which it would advantageous to do so.

Quoting from the manual :

The alloca() function is machine and compiler dependent. On many
systems its implementation is
buggy. Its use is discouraged.

The more serious issue:
On many systems alloca() cannot be used inside the list of arguments
of a function call, because
the stack space reserved by alloca() would appear on the stack in the
middle of the space for the
function arguments.