In article <2525643717796095.wa.m42tomibmmainyahoo....@listserv.ua.edu> you wrote: > On Wed, 27 Jan 2016 16:31:17 -0500, Farley, Peter wrote:
> >Some of us may have to be dragged kicking and screaming into > >that 64-bit future because of PDSE-fear, but it is coming nonetheless. > There is another impediment, IMO. That is that XPLINK-64 cannot easily > coexist with either XPLINK-31 or with standard linkage. So, if you have > an application that consists of several modules, they have to be either > all XPLINK-64 or none of them. > -- > Tom Marchant Your third sentence is refuted by your second. Yes, it's not easy, but it can be done. IBM could make this easier if they would do the work to provide 64-bit DLLs for more of their products. As more customers go to 64-bit, the presure for that will build. Another issue is IEEE float. Hex float can be used by XPLINK callers, but the HFP library is non- XPLINK, so ends up being much slower. If your code is in a high-level language, you can mostly get by by using FLOAT(IEEE), but if you have lots of assembler math routines (I wonder who would have those?), you're going to have to dig in a bit. I highly recommend the floating point chapters in John Erman's assembler book for teaching yourself the fine points of this stuff. The pop is probably good enough for the mathmeticians among us, but I needed a primer. You can download here: http://idcp.marist.edu/enterprisesystemseducation/Assembler%20Language%20Programming%20for%20IBM%20z%20System%20Servers.pdf -- Don Poitras - SAS Development - SAS Institute Inc. - SAS Campus Drive sas...@sas.com (919) 531-5637 Cary, NC 27513 ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
