Hi Pieter,
Yes I did figure out that to use of MF=E/L or SF=E/L will help to avoid the
SIIS scenario for IBM Macros.
Thx much for your input have a good day Philippe
For IBM macros that do SIIS, use the MF=E/L or SF=E/L variants.
-Original Message-
From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On
Behalf Of Philippe Cloarec
Sent: 14 November 2016 07:06
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: SIIS "issue&q
Hi John,
Thx much for your input. Yes, I was planning to use newer instructions but I
did realize seeing current design of the applications I am reviewing this will
generate some extra needed time to be sure all is ok and I cannot afford this
in current context. To implement Baseless pgm or
)
From: Philippe Cloarec <philippe.cloa...@gmail.com>
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Date: 12/11/2016 21:16
Subject: Re: SIIS "issue" after upgrade to z13 machine.
Sent by:IBM Mainframe Assembler List
<ASSEMBLER-LIST@LISTSERV.UGA.EDU>
Hi Marti
Apparently this is needed for some...sure SIIS is NOT new for sure, BUT
performance issue was obviously revealed after upgrade
to a z13, please do read documents referenced in previous updates. Have a good
evening.
ASSEMBLER-LIST@LISTSERV.UGA.EDU
> Subject: Re: SIIS "issue" after upgrade to z13 machine.
>
> Hi Rob and Martin,
> Thx for your input. MY primary purpose is to detect SIIS scenarios across
> code I have to scan and to fix them. Code is not complicated I mean you do
&
Hi Rob and Martin,
Thx for your input. MY primary purpose is to detect SIIS scenarios across code
I have to scan and to fix them. Code is not complicated I mean you do have
basic i/o processing against files and some characters and numerics fields are
processed.
There are some Ex(ecute)
On 12 November 2016 at 10:10, Philippe Cloarec
wrote:
>
> Since we do talk of CPU cycles savings here I will check for AGI cases and
> their resolution and try to implement instruction grouping as much I can.
>
> From my humble point this is a real topic and all z13
Pieter,
>> The would change most branch instructions to relative branches.
I say not "most" but "all".
Here is a construct that IEABRC (or its X brother) does not handle
well
B *+4*NUMACONS
DC AL4(SOMEWHERE)
DC AL4(OVER_THE_RAINBOW)
.
.
but these are rare and
Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On
Behalf Of Philippe Cloarec
Sent: 12 November 2016 11:10
To: ASSEMBLER-LIST@LISTSERV.UGA.EDU
Subject: Re: SIIS "issue" after upgrade to z13 machine.
Hi Martin and Rob,
First thx much for your input. Here are 3 links wh
Hi Martin and Rob,
First thx much for your input. Here are 3 links which relate the "issue" and
common code for which we fall into a SIIS scenario:
https://www.google.fr/#q=istream_flash_062606_v4
Rob,
excellent example that would be obvious when converting to
"baseless" (and eliminated to be performance-perfect even without going
to RENT)
Martin
Philippe,
>> ... no time to rewrite them as RENT.
Well - if this is the case then i would not invest into using newer
instructions (with the exception of the fairly old "relative &
immediate" instructions)
You need to understand the code to replace instruction sequences.
To make code RENT is
The example that I have seen was where the customer had a linkage model for
small subroutines that used a static save area and local storage after a
branch at the start of the program. That's painful for small routines
because a lot of the code is in the same cache line and gets hit each time
you
Hi,
As you may know, there is some kind of performance issue because of SIIS(Store
Into Instruction Stream) after upgrade to a z13 machine in some scenarios. CPU
increase of 30% can be seen in some case, so it may be good to perform to
related changes to avoid issue from reoccuring. Did start
15 matches
Mail list logo