There seems to be a lot of confusion about how various header files are
included in compiles. I am certainly not an expert in GCC on VM, C
library options, or CMS itself.
But, the discussion begs the question: Is there a complete list of
which disks / modules / etc that documents the _present_ structure?
Everyone seems to see a piece of the problem not the entire structure.
Such a list might point to how it could be improved.
Just a thought,
Harold Grovesteen
toggle quoted message
Show quoted text
On Tue, 2021-12-07 at 10:06 +0000, Dave Wade wrote:
Mark,
Well the mess is there BECAUSE there are two sets of ¡°.h¡± files and
CMS does not have a directory structure.
So the production GCC EXEC accesses the GCCCMS disks that contain the
¡°.h¡± files that match the library you have selected PDPCLIB or GCCLIB
There are two ways to fix this. One would be to change the GCC
compiler so you could specify which disks it searched for ¡°.h¡± files.
The other would be to modify the compiler so it could read ¡°.h¡± files
from emulated OS Partitioned Data Sets as I think the IBM compiler
does.
Neither seem attractive, but what I would think could be done is to
fix the help that¡¯s returned by ¡°HELP GCC (MORE¡± to explain what is
happening and why.
Its needs fixing anyway, its also wrong:-(
Dave
Groups.io Links:
You receive all messages sent to this group.
View/Reply Online (#3680) | Reply To Group | Reply To Sender | Mute
This Topic | New Topic
Mute #vmce
Your Subscription | Contact Group Owner | Unsubscribe [
h.grovsteen@...]
_._,_._,_
