inline ss_sint8_t &transfer_resource(ss_sint8_t &r, ss_sint8_t v)               { return (r = v); }
inline ss_uint8_t &transfer_resource(ss_uint8_t &r, ss_uint8_t v)               { return (r = v); }
inline ss_sint16_t &transfer_resource(ss_sint16_t &r, ss_sint16_t v)            { return (r = v); }
inline ss_uint16_t &transfer_resource(ss_uint16_t &r, ss_uint16_t v)            { return (r = v); }
inline ss_sint32_t &transfer_resource(ss_sint32_t &r, ss_sint32_t v)            { return (r = v); }

Changing it to

    { return ((r = v), r); }

fixes it, but it still seems like a bug to me.

Other compilers don't have a problem with the original form

Any movement on this?

"Matthew" <admin.hat@stlsoft.dot.org> wrote in message news:cvfag8$2vl2$1@digitaldaemon.com...
> inline ss_sint8_t &transfer_resource(ss_sint8_t &r, ss_sint8_t v)
{ return (r = v); }
> inline ss_uint8_t &transfer_resource(ss_uint8_t &r, ss_uint8_t v)
{ return (r = v); }
> inline ss_sint16_t &transfer_resource(ss_sint16_t &r, ss_sint16_t v)
{ return (r = v); }
> inline ss_uint16_t &transfer_resource(ss_uint16_t &r, ss_uint16_t v)
{ return (r = v); }
> inline ss_sint32_t &transfer_resource(ss_sint32_t &r, ss_sint32_t v)
{ return (r = v); }
>
> Changing it to
>
>     { return ((r = v), r); }
>
> fixes it, but it still seems like a bug to me.
>
> Other compilers don't have a problem with the original form
>
>