On 30/10/2020 12:58 am, Paul Gilmartin wrote:
I suspect he's confusing "address space" with "execution environment":
https://pubs.opengroup.org/onlinepubs/9699919799/utilities/V3_chap02.html#tag_18_12
If you're referring to me: it's possible, but I don't think so.
Referring to the comparison of bash and /bin/sh I linked previously, I
can see the address space from the job number field in SMF.
The main bash process runs in STC06733 (a BPX initiator address space).
If I look at e.g the setfacl command I can see STC06734 (i.e. a
different address space) initially runs bash, but then sub step 0 ends
and sub step 1 starts, running the setfacl command.
That appears to be a fork of bash followed by exec of setfacl in a new
sub step as described here:
https://www.ibm.com/support/knowledgecenter/SSLTBW_2.4.0/com.ibm.zos.v2r4.ieag200/ext.htm
If I compare /bin/sh:
The top level SH is running in STC06733.
setfacl runs in STC06736, sometimes STC06737. It is sub step 0 - there
is no sub step for the shell.
There is one SH step for each iteration of the loop - I think this is
the pipe.
The mkdir steps seen under bash disappear completely - I guess it is
built into the shell.
When I tried with _BPX_SHAREAS all the individual steps for the shell
commands disappeared, except for the SH step
--
Andrew Rowley
Black Hill Software
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
