On Sat, 7 Jan 2017 23:19:13 +0200, Binyamin Dissen <[email protected]> 
wrote:

>On Sat, 7 Jan 2017 16:57:05 +0000 "Blaicher, Christopher Y."
><[email protected]> wrote:
>
>:>Generally speaking, a page sized request will return a zeroed page, but 
>there is no guarantee for that unless you specify that on the GETMAIN.
>
>What is the case where it would not return a zeroed page?

The case specified by the OP would do it. He hypothesized doing a GETMAIN for 
4096 bytes and getting storage at address 01B00C00. As that is not on a page 
boundary, I think it could only occur if someone had done a GETMAIN for, e.g., 
8192 bytes, gotten the storage from 01B00000 through 01B01FFF, then freed 4096 
bytes starting at 01B0C00. Whatever the application left at that address before 
the FREEMAIN would still be there when the GETMAIN returned that same, 
previously used storage.

Nothing guarantees that a 4096-byte request will end up returing storage on a 
page boundary, unless you specify that on the request.

-- 
Walt

----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to