Making a function safe for use in an ISR

In article <m3ireledq6.fsf@Claudio.Messina>, Mark L Pappin <mlp@acm.org> 
writes
>Chris Hills <chris@phaedsys.org> writes:
>
>> Paul Keinanen <keinanen@sci.fi> writes
>>> "Gary Pace" <xxx@yyy> wrote:
>>>> - i is signed, int may be 8,16,32 (maybe 64?) bits in size
>>> If this is supposed to be standard C, then int can not be 8 bits.
>
>> No,  this  is for an embedded MCU so it can be 8 bits.
>
>Not if you want to call the language C.
>You can call it "C-like" if you want,

You are being a pedant and pointless.

> but for the last 17 years there
>has been an international standard defining the language called C

I know I have been on that panel for about a decade.

>and
>one of its most fundamental features is that the type 'int' can hold
>all values between 32767 and -32767 inclusive.

This is not correct. Type in should be at least 16 bits. Ints can 
legally be any size above 16 bits. There are systems with 64 bit ints.

Also an int can be signed or unsigned. An unsigned int does not hold 
negative values...

Of course both signed and unsigned char are integer types though a char 
does not have to be 8 bits. Hence the communications industry having the 
type "octet" which we have suggested as a standard type on more than one 
occasion.

>A language with a type
>'int' that can not hold at least this range can not be called C.

As I said you are a pedant and pointless. The difference between theory 
and practice....

There are many dialects of C Most break the rules somewhere. By your 
definition there are only , I think, 6 C compilers world wide but they 
only could for a minute fraction of the "c" programming in the world. 
The vast majority do not fully implement ISO 9899:1999 but rather 
something between C95 and C99 *with extensions*

There are some compilers that are as compliant as most of  the other c 
compilers out there that for reasons of hardware architecture and 
efficiency my have 8 bit ints.  It is not common  but I would not stop 
them being called C compilers for just that reason alone.

>The C standard _does_ allow for various extensions, and makes the
>distinction between "freestanding" and "hosted" implementations (most
>embedded platforms will be the former, which for instance doesn't need
>to provide all or even any of the standard library) but reducing the
>range of _the_ basic type in this way is just not on.
>mlp

As I said all "c" compilers have extensions and restrictions on 
ISO9899:1999.  I think only 6 fully implement the standard, all the 
common ones don't, ALL of the C compilers have extensions.

C is like English, rarely ever used in it's pure form.



-- 
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
/\/\/ chris@phaedsys.org      www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

Chris Hills wrote:
> Mark L Pappin <mlp@acm.org> writes
>> Chris Hills <chris@phaedsys.org> writes:
>>> Paul Keinanen <keinanen@sci.fi> writes
>>>
... snip ...
>>>
>>>> If this is supposed to be standard C, then int can not be 8 bits.
>>>
>>> No,  this  is for an embedded MCU so it can be 8 bits.
>>
>> Not if you want to call the language C.
>> You can call it "C-like" if you want,
> 
> You are being a pedant and pointless.
> 
>> but for the last 17 years there
>> has been an international standard defining the language called C
> 
> I know I have been on that panel for about a decade.

Then you should know better.  I suspect you are largely ignored.

> 
>> and one of its most fundamental features is that the type 'int'
>> can hold all values between 32767 and -32767 inclusive.
> 
> This is not correct. Type in should be at least 16 bits. Ints can
> legally be any size above 16 bits. There are systems with 64 bit ints.

All of which can hold values between 32767 and -32767 as he said.

> 
> Also an int can be signed or unsigned. An unsigned int does not hold
> negative values...
> 
> Of course both signed and unsigned char are integer types though a
> char does not have to be 8 bits. Hence the communications industry
> having the type "octet" which we have suggested as a standard type
> on more than one occasion.

unsigned ints are not ints.  They are a different type and follow
different rules, such as the use of modulo arithmetic.  Which
doesn't affect the statement about ints.

> 
>> A language with a type
>> 'int' that can not hold at least this range can not be called C.
> 
> As I said you are a pedant and pointless. The difference between
> theory and practice....
> 
> There are many dialects of C Most break the rules somewhere. By
> your definition there are only , I think, 6 C compilers world
> wide but they only could for a minute fraction of the "c" 
> programming in the world. The vast majority do not fully implement
> ISO 9899:1999 but rather something between C95 and C99 *with
> extensions*

Then it isn't a C compiler, by definition.  It may still be
useful.  The point of a standard is to allow confidence that source
code will work on any system that is standards compliant.  Use of
extensions should be limited to clearly identified modules.

-- 
 <http://www.cs.auckland.ac.nz/~pgut001/pubs/vista_cost.txt>

 "A man who is right every time is not likely to do very much."
                           -- Francis Crick, co-discover of DNA
 "There is nothing more amazing than stupidity in action."
                                             -- Thomas Matthews

In article <45C33455.7887F4C3@yahoo.com>, CBFalconer 
<cbfalconer@yahoo.com> writes
>Chris Hills wrote:
>> There are many dialects of C Most break the rules somewhere. By
>> your definition there are only , I think, 6 C compilers world
>> wide but they only could for a minute fraction of the "c"
>> programming in the world. The vast majority do not fully implement
>> ISO 9899:1999 but rather something between C95 and C99 *with
>> extensions*
>
>Then it isn't a C compiler, by definition.

This sort of stupidity is prevalent on comp.lang.c but thankfully not on 
the standards panels. I am surprised to see in an embedded NG.

By your definition there are only 6 C compilers and non of them in the 
embedded area.
-- 
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
/\/\/ chris@phaedsys.org      www.phaedsys.org \/\/\
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/