I am finding this interesting. One thing I think you are missing was the advisability to use STMG/LMG to save and restore all 64 bits of the registers. Not a lot of stuff running above the bar as yet, but it will be there sooner than we expect. :)
> -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of Leonardo Vaz > Sent: Tuesday, February 16, 2016 11:12 AM > To: [email protected] > Subject: Re: CSVFETCH exit > > I see. In the case of CSVFETCH there is a 1024-byte work-area passed, so no > getmain should be needed, in this case, STM should be the best option? > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of Blaicher, Christopher Y. > Sent: Tuesday, February 16, 2016 1:50 PM > To: [email protected] > Subject: Re: CSVFETCH exit > > BAKR/PR does take a lot more time then STM/LM, but most times you can't > just use STM/LM, you also have a GETMAIN/FREEMAIN for a register save > area and other setup/restore things to do. > > So, once you factor in those other overheads, which is faster? I don't know > as > I haven't set up the tests to validly compare the different methods possible. > I > have seen several different ways to avoid the GETMAIN/FREEMAIN > overheads. > > Chris Blaicher > Technical Architect > Software Development > Syncsort Incorporated > 50 Tice Boulevard, Woodcliff Lake, NJ 07677 > P: 201-930-8234 | M: 512-627-3803 > E: [email protected] > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of Leonardo Vaz > Sent: Tuesday, February 16, 2016 1:41 PM > To: [email protected] > Subject: Re: CSVFETCH exit > > Well, I guess using the linkage stack is cleaner because the state of the > caller > is preserved. > > Thanks for your reply, that is what I wanted to know, I was going to use STM > but I remember reading somewhere before (maybe linkedin?) that they would > never approve any assembler code written in the past 20 years starting with > STM R14,R12,12(R13). I just would like to hear someone that would prefer > BAKR and why. > > Thanks, > Leo > > -----Original Message----- > From: IBM Mainframe Discussion List [mailto:[email protected]] > On Behalf Of Tom Marchant > Sent: Tuesday, February 16, 2016 12:53 PM > To: [email protected] > Subject: Re: CSVFETCH exit > > On Tue, 16 Feb 2016 16:52:03 +0000, Leonardo Vaz wrote: > > >I apologize if the question is silly, but I am wondering if for a > >performance sensitive exit like this one I should use STM/LM instead of > >BAKR/PR. I believe it's "cheaper" to do STM/LM, bur "cleaner" to do BAKR, > right? > > Why do you think BAKR is "cleaner"? > > BAKR/PR is MUCH slower than STM/LM (or STMG/LMG if you are using the > high halves of any of registers 2-12). > > -- > Tom Marchant > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send email to > [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send email to > [email protected] with the message: INFO IBM-MAIN > > ________________________________ > > > > ATTENTION: ----- > > The information contained in this message (including any files transmitted > with this message) may contain proprietary, trade secret or other confidential > and/or legally privileged information. Any pricing information contained in > this message or in any files transmitted with this message is always > confidential and cannot be shared with any third parties without prior written > approval from Syncsort. This message is intended to be read only by the > individual or entity to whom it is addressed or by their designee. If the > reader > of this message is not the intended recipient, you are on notice that any use, > disclosure, copying or distribution of this message, in any form, is strictly > prohibited. If you have received this message in error, please immediately > notify the sender and/or Syncsort and destroy all copies of this message in > your possession, custody or control. > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send email to > [email protected] with the message: INFO IBM-MAIN > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, send email to > [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
