Re: [Gm2] ADR of ADDRESS

[SiTex Graphics]

Hi Gaius,

Thank you very much for this fix.  The application now works properly
in all my tests so far.  This tool is built from a subset of our main
code base, so I'm one step closer to getting the main application to
build.

Thanks again,
Scott


On Tue, Dec 1, 2009 at 7:14 PM, Gaius Mulley <address@hidden> wrote:
> SiTex Graphics <address@hidden> writes:
>
>> Hi Gaius,
>>
>> Attached is a test file illustrating a problem with ADR() applied to a
>> variable holding an address.  Compile with
>>
>> gm2 -fiso -o adraddress -fmakeall adraddress.mod
>>
>> The executable generates an invalid location exception.  ADR appears
>> to return the contents of the address variable instead of the address
>> of the variable itself...if that makes sense.
>>
>> If you're wondering why someone would want to do this, we have an
>> interface to an external library that expects a string to be passed as
>> a pointer to a pointer to the string.
>>
>> More progress:  another of our tools now compiles and some of its
>> functionality appears to work.  Have to wait for this bug fix to know
>> if it's all working.
>
> Hi Scott,
>
> I believe this bug is now fixed in the cvs - and all regressions pass.
> Will be interested to hear about the success/failure of your
> application !
>
> regards,
> Gaius
>