Hello again list! 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?
Thanks, Leo -----Original Message----- From: IBM Mainframe Discussion List [mailto:[email protected]] On Behalf Of Tom Brennan Sent: Friday, February 12, 2016 12:50 PM To: [email protected] Subject: Re: CSVFETCH exit I agree, unsafe. But I could probably argue that just letting a system programmer logon is unsafe :) I had various checks to try to prevent the hooks from causing issues, including (if I remember correctly) using the LRA instruction to check for a valid address in ECSA prior to access. I didn't want to cause an 0C4 in someone's SVC 8 just because I messed up. Over the 4 years or so the hooks were in place I was kind of proud they had never caused problems or crashed a system. And they had to co-exist with various vendor product hooks on the same SVC's. That work gave me a good look at what the poor vendors have to go through when no better interface exists. Leonardo Vaz wrote: > More difficult indeed, but at least it is now a documented programming > interface, which is a huge jump from having to play around with > arguably unsafe hooks :) ---------------------------------------------------------------------- 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
