Date: Mon, 26 Jan 1998 18:54:21 -0800 (PST)
Message-Id: <199801270254.SAA29226@adit.ap.net>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: "d-range!" <d-range AT thefridge DOT et DOT fnt DOT hvu DOT nl>, djgpp AT delorie DOT com
From: Nate Eldredge <eldredge AT ap DOT net>
Subject: Re: WATCOM #pragma to DJGPP __asm__ ?!!
Precedence: bulk

At 10:31  1/26/1998 +0100, d-range! wrote:
>Nate Eldredge wrote:
>
>> >My question: how can I make inline functions like this one to work with
>> >DJGPP. I read the DJGPP assembler tutorial, and I know you can #define the
>> >function like this,
>> >
>> >#define FixSHR(arg1,arg2,arg3) __asm__ \
>> >            "sarl %1,%0" \
>> >       : "=r" (arg3) \
>> >       : "0" (arg1), "1" (arg2) \
>> >       : "0";
>> Try this:
>> inline int FixSHR(int n, unsigned c)
>> {
>>    int result;
>>    asm ("sarl %2,%0"
>>         : "=g" (result)
>>         : "0" (n), "cI" (c));
>>    return result;
>> }
>
>Does this produce *inline* assembly? It looks like if the compiler makes a CALL
>for every invocation.
Yes, if you include the code in each source file which uses it and compile
with `-O'.
See the info page "gcc" "C Extensions" "Inline" for more info on inline
functions. The `extern inline' syntax may be particularly helpful here.

>> Incidentally, since your operands are `int', `FixSHR(a,b)' has the same
>> effect as `a >> b'.
>
>Ah... WATCOM always does SHR instead of SAR
That's silly. A signed value should get an arithmetic shift.

Nate Eldredge
eldredge AT ap DOT net