None.
Enables compiler support for decimal floating-point types and literals.
-qnodfp
If you enable -qdfp for a -qarch value that does not support decimal floating-point instructions, -qfloat=dfpemulate is automatically enabled, and the decimal floating-point operations are performed by software. This may cause a slowdown in the application's runtime performance.
Note that runtime support for decimal floating-point operations is available only on AIX® for POWER® version 5.3 with the 5300-06 Technology Level or later. If you enable -qdfp on a version of the operating system that does not provide runtime support, your application will compile, but it may not link or run.
Programs that use decimal floating-point functions or macros defined in the <math.h> include file should not be compiled on AIX 5.2 or on older levels of AIX 5.3 or 5.4. because those functions and macros won't be retrofitted into 5.2.
When -qdfp is in effect, __IBM_DFP__ is predefined to a value of 1; otherwise it is undefined.