We have ended up with many of these in our code because a senior
developer felt that too many single-usage labels cluttered the code and
made it less readable. (He also required that a blank line follow every
branch statement and precede the target of every branch, whether labeled
or not).
The concession we got from him was to identify every instruction
bypassed (and yes, there were multiple instructions in some cases) by
using, for example, Bx *+4+2+6+4. Now some of these have become Jx *+....
These haven't been a maintenance problem. Usually when maintenance is
applied affecting these constructs a label is added as the B is changed
to a J.
Gary Weinhold
DKL
On 2015-07-02 19:37, zMan wrote:
Yeah, OK, I don't know why I was taught to do this, but I've sort of become
suspicious that it was Evil. I will stop. Or try to. Maybe there's a
support group...
On Thu, Jul 2, 2015 at 7:13 PM, Bob Rutledge <[email protected]> wrote:
On 7/2/2015 6:21 PM, Paul Gilmartin wrote:
On 2015-07-02 15:50, [email protected] wrote:
I admit I am in the minority here.
For a single instruction I have no problems with a B *+4
But thats just my Opinion
Is that, like, IEFBR14 is a whole lot faster than IEFBR15?
The fastest branch is the one not taken.
Bob
