Re: QPROCT and QPRODT

On 01/05/2010, at 10:54 AM, Richard Thomas wrote:

"MXD is the same as POS and HI" - Agreed, when you look at the generated
code within a program. I'm not disputing this.

I naively expected QPRCRTPG to enforce instruction-specific synonyms for the
indicator/branch forms in the IRP (e.g. POS on ADDN is OK, but MXD and HI
aren't).

My guess is the parser doesn't care. It verifies the op-code against the QPROCT entries (name and optional forms), checks the number of parameters and their types against QPROCD, then converts the condition text to a bit value and compares (ANDs) that to the bits in flag bytes 2 (positive) or 3 (negative). It apparently doesn't bother to check the mnemonic condition itself makes sense. Not entirely surprising since the generated code is the same. If you as the developer want to specify HI when POS might make more sense why should the compiler force you to do so when the result is the same?

Anyway, thanks for reporting the condition information.

Regards,
Simon Coulter.
--------------------------------------------------------------------
FlyByNight Software OS/400, i5/OS Technical Specialists

http://www.flybynight.com.au/
Phone: +61 2 6657 8251 Mobile: +61 0411 091 400 /"\
Fax: +61 2 6657 8251 \ /
X
ASCII Ribbon campaign against HTML E-Mail / \
--------------------------------------------------------------------