Csect - Dsect Question
I have a need to create message table, with the following attributes: 1. MSGID = 9 chars 2. Length of msg 3. Message I would like this "tab;e" in loose terms to be external. I have never done external dsects. Am in right i can do that , create a external message table, bring it into storage and index into it to find messages ? If so where can I find a example .. Regards, Scott
Re: Csect - Dsect Question
On 3/30/2016 7:05 AM, Scott Ford wrote: I have a need to create message table, with the following attributes: 1. MSGID = 9 chars 2. Length of msg 3. Message I would like this "tab;e" in loose terms to be external. I have never done external dsects. Am in right i can do that , create a external message table, bring it into storage and index into it to find messages ? If so where can I find a example .. For example: MsgEntry DSECT , MsgIdDSCL9 MsgSize DSXL1 MsgText DSCL80 MsgLenth EQU *-MsgEntry Code LOCTR ,Switch to code ... LOAD EPNAME==CL8'MSGTABLE'Load external message table NILH R1,X'7FFF' Turn off AMODE indicators NILL R1,X'FFFE' (same) STG R1,MsgTable Save clean table address ... LLGH R4,MsgNumGet message number AGHI R4,-1Make relative to zero MGHI R4,MsgLenth Compute offset into table AGR4,MsgTable Point to message entry USING MsgEntry,R4 http://www.phoenixsoftware.com/
Re: Csect - Dsect Question
Wow, Ed how do I say thank you, and yep I am a following your code. It's great... Regards, Scott On Wednesday, March 30, 2016, Ed Jaffe wrote: > On 3/30/2016 7:05 AM, Scott Ford wrote: > >> I have a need to create message table, with the following attributes: >> >> 1. MSGID = 9 chars >> 2. Length of msg >> 3. Message >> >> I would like this "tab;e" in loose terms to be external. I have never done >> external dsects. Am in right i can do that , create a external message >> table, bring it into >> storage and index into it to find messages ? If so where can I find a >> example .. >> > > For example: > > MsgEntry DSECT , > MsgIdDSCL9 > MsgSize DSXL1 > MsgText DSCL80 > MsgLenth EQU *-MsgEntry > > Code LOCTR ,Switch to code > ... > LOAD EPNAME==CL8'MSGTABLE'Load external message table > NILH R1,X'7FFF' Turn off AMODE indicators > NILL R1,X'FFFE' (same) > STG R1,MsgTable Save clean table address > ... > LLGH R4,MsgNumGet message number > AGHI R4,-1Make relative to zero > MGHI R4,MsgLenth Compute offset into table > AGR4,MsgTable Point to message entry > USING MsgEntry,R4 LAY R1,OutArea Point to output area > MVC 0(L'MsgId,R1),MsgId Copy message id > MVI L'MsgId(R1),C' ' Set blank separator > LAR15,L'MsgID+1(,R1) Point where MsgText goes > LLGC R14,MsgSize Get size of message > AGHI R14,-1 Make relative to zero > EXR14,MoveTxt Copy the message text > > Constant LOCTR , Switch to constants > MoveTxt MVC 0(*-*,R15),MsgText Dummy for execute > Code LOCTR , Switch back to code > > LAR0,1(R14,R15)Compute total msg length > SGR R0,R1 in R0 for WriteMsg service. > JAS R14,WriteMsg R1 = TextPtr, R0 = TextgLen > DROP R4 ... > > -- > Edward E Jaffe > Phoenix Software International, Inc > 831 Parkview Drive North > El Segundo, CA 90245 > http://www.phoenixsoftware.com/ >
Re: Csect - Dsect Question
On Wed, 30 Mar 2016 07:42:10 -0700, Ed Jaffe wrote: > LOAD EPNAME==CL8'MSGTABLE'Load external message table > NILH R1,X'7FFF' Turn off AMODE indicators > NILL R1,X'FFFE' (same) > STG R1,MsgTable Save clean table address ITYM R0. And the manual doesn't specify that the address returned is a clean 64-bit address except if it is AMODE 64. So I'd suggest replacing the NILH with LLGTR R0,R0 -- Tom Marchant
Re: Csect - Dsect Question
On 3/31/2016 2:03 PM, Tom Marchant wrote: On Wed, 30 Mar 2016 07:42:10 -0700, Ed Jaffe wrote: LOAD EPNAME==CL8'MSGTABLE'Load external message table NILH R1,X'7FFF' Turn off AMODE indicators NILL R1,X'FFFE' (same) STG R1,MsgTable Save clean table address ITYM R0. And the manual doesn't specify that the address returned is a clean 64-bit address except if it is AMODE 64. So I'd suggest replacing the NILH with LLGTR R0,R0 Haha! You're right (re: R0) and good suggestion. That's better code... (Still not a bad effort for typing HLASM code into an email. ;-) ) Another alternative would be to make MsgTable a fullword and change the AG to an AGF in the indexing/lookup code. -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 http://www.phoenixsoftware.com/
Re: Csect - Dsect Question
LLGT does not zero the 64-bit mode bit (63). On 03/31/2016 11:19 PM, Ed Jaffe wrote: On 3/31/2016 2:03 PM, Tom Marchant wrote: On Wed, 30 Mar 2016 07:42:10 -0700, Ed Jaffe wrote: LOAD EPNAME==CL8'MSGTABLE'Load external message table NILH R1,X'7FFF' Turn off AMODE indicators NILL R1,X'FFFE' (same) STG R1,MsgTable Save clean table address ITYM R0. And the manual doesn't specify that the address returned is a clean 64-bit address except if it is AMODE 64. So I'd suggest replacing the NILH with LLGTR R0,R0 Haha! You're right (re: R0) and good suggestion. That's better code... (Still not a bad effort for typing HLASM code into an email. ;-) ) Another alternative would be to make MsgTable a fullword and change the AG to an AGF in the indexing/lookup code.
Re: Csect - Dsect Question
The complete "corrected" version of the loading is this .. LOAD EPNAME==CL8'MSGTABLE'Load external message table LLGTR R0,R0clean high order word and low high order bit NILL R0,X'FFFE' (same) STG R0,MsgTable Save clean table address ... -- Martin Pi_cap_CPU - all you ever need around MWLC/SCRT/CMT in z/VSE more at http://www.picapcpu.de
Re: Csect - Dsect Question
If we're insisting on correctness (I wasn't, since Ed just blasted code from his head into an email), LOAD has no EPNAME argument. Coded as below, EPNAME should be EPLOC. On 2016-04-01 03:14, mar...@pi-sysprog.de wrote: The complete "corrected" version of the loading is this .. LOAD EPNAME==CL8'MSGTABLE'Load external message table LLGTR R0,R0clean high order word and low high order bit NILL R0,X'FFFE' (same) STG R0,MsgTable Save clean table address ... -- Regards, Gord Tomlin Action Software International (a division of Mazda Computer Corporation) Tel: (905) 470-7113, Fax: (905) 470-6507
Re: Csect - Dsect Question
On 4/1/2016 7:50 AM, Gord Tomlin wrote: If we're insisting on correctness (I wasn't, since Ed just blasted code from his head into an email), LOAD has no EPNAME argument. Coded as below, EPNAME should be EPLOC. On 2016-04-01 03:14, mar...@pi-sysprog.de wrote: The complete "corrected" version of the loading is this .. LOAD EPNAME==CL8'MSGTABLE'Load external message table LLGTR R0,R0clean high order word and low high order bit NILL R0,X'FFFE' (same) STG R0,MsgTable Save clean table address ... Thanks, Gord! :-[ Obviously I should have "blasted" the code from my head into a source code editor, wrapped JCL around it, assembled it, executed and debugged it, and then finally pasted it into an email... LOL ;-) -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 http://www.phoenixsoftware.com/
Re: Csect - Dsect Question
Ed, >> ...and then finally pasted it ... cann't you see the power of a "mass"-compile any bug left we have yet to find? -- Martin Pi_cap_CPU - all you ever need around MWLC/SCRT/CMT in z/VSE more at http://www.picapcpu.de
Re: Csect - Dsect Question
<<<any bug left we have yet to find? If you intend to use RMODE64 in the future with that code you will want to NOT use that LLGTR. Chuck Arney Arney Computer Systems Web: http://zosdebug.com Facebook: http://www.facebook.com/arneycomputer -Original Message- From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On Behalf Of mar...@pi-sysprog.de Sent: Friday, April 01, 2016 12:47 PM To: ASSEMBLER-LIST@LISTSERV.UGA.EDU Subject: Re: Csect - Dsect Question Ed, >> ...and then finally pasted it ... cann't you see the power of a "mass"-compile any bug left we have yet to find?
Re: Csect - Dsect Question
Chuck, holy sh.. is RMODE64 supported yet on any op-sys (and don't forget Scott; he wants it for COBOL)? -- Martin Pi_cap_CPU - all you ever need around MWLC/SCRT/CMT in z/VSE more at http://www.picapcpu.de
Re: Csect - Dsect Question
z/TPF has supported RMODE64 code for over 10 years. But we're a little different than most other OSes. :-) - mb IBM Mainframe Assembler List wrote on 04/01/2016 02:17:06 PM: > is RMODE64 supported yet on any op-sys (and don't forget > Scott; he wants it for COBOL)?
Re: Csect - Dsect Question
It works fine on z/OS 1.13 and above. We have lots of code running from there. However, only a directed LOAD can be done into RMODE64 or you have to do the program management yourself which is what we do. I wrote a blog post on running RMODE64 code a while back. Chuck Arney Arney Computer Systems Web: http://zosdebug.com Facebook: http://www.facebook.com/arneycomputer -Original Message- From: IBM Mainframe Assembler List [mailto:ASSEMBLER-LIST@LISTSERV.UGA.EDU] On Behalf Of mar...@pi-sysprog.de Sent: Friday, April 01, 2016 1:17 PM To: ASSEMBLER-LIST@LISTSERV.UGA.EDU Subject: Re: Csect - Dsect Question Chuck, holy sh.. is RMODE64 supported yet on any op-sys (and don't forget Scott; he wants it for COBOL)? -- Martin Pi_cap_CPU - all you ever need around MWLC/SCRT/CMT in z/VSE more at http://www.picapcpu.de
Re: Csect - Dsect Question
Mark, >> z/TPF has supported RMODE64 code for over 10 years. Oups- I love writing code transformed to being run on the metal by HLASM (I do know Pops by heart), but yet my CV lacks doing anything in TPF. -- Martin Pi_cap_CPU - all you ever need around MWLC/SCRT/CMT in z/VSE more at http://www.picapcpu.de
Re: Csect - Dsect Question
On Fri, 1 Apr 2016 12:55:44 -0500, Chuck Arney wrote: >If you intend to use RMODE64 in the future with that code you will want to >NOT use that LLGTR. Right. To cover that case you might want something like this: LOAD EP=MSGTABLE Load external message table TMLL R0,X'0001' RMODE(64) JORMODE64 LLGTR R0,R0clean high order word and low high order bit RMODE64 NILL R0,X'FFFE' (same) STG R0,MsgTable Save clean table address -- Tom Marchant
Re: Csect - Dsect Question
On 3/31/2016 2:03 PM, Tom Marchant wrote: ITYM R0. Indeed! And the manual doesn't specify that the address returned is a clean 64-bit address except if it is AMODE 64. So I'd suggest replacing the NILH with LLGTR R0,R0 Empirical testing shows R0 is returned with a clean 64-bit address, even for an old 24-bit program: TCB#6 RB#1 - z/XDC TPUT INTERFACE XDC ===> _ _0E957100 0s (A.S.EDJX2) --- EJESSUBS.EJESSUBS+2100, EJESSUBS+2100, XPRIVATE+1A57100 _ +2100 *** _ +2100* Empirical Test for My ASSEMBLER-LIST Friends* _ +2100 *** _ +2100 EBEF C060 0004 LMG R14,R15,=4FD'-1' Put garbage in R14-R1 _ +2106 LOAD EPLOC==CL8'IEFBR14' Load a module _ +2106 4100 C0C0 + LA0,=CL8'IEFBR14' LOAD PARAMETER INTO REG 0 _ +210A 1B11 + SR1,1 SHOW NO DCB PRESENT _ +210C 0A08 + SVC 8 _ +210E > DCH'0' Force an abend _ +210E > +0 *..* XDC ===> L RWREGS _ RW0 _00EC5000 _0001 _7F363434 _ _ RW4 _7F33A71E _0E9747CE _0041 _F000 _ RW8 _7F466D00 _7F33C1A4 _7F33A000 _0E955000 _ RW12 _0E98E7B0 _7F466E00 _ 0001_ -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 http://www.phoenixsoftware.com/
Re: Csect - Dsect Question
On Fri, 1 Apr 2016 15:08:59 -0400, Tom Marchant wrote: >TMLL R0,X'0001' RMODE(64) >JORMODE64 Oops. I mean TMLL R0,X'0001' AMODE(64) JOAMODE64 Yes, leave bits 0-32 as is -- Tom Marchant
Re: Csect - Dsect Question
On Fri, 1 Apr 2016 12:36:01 -0700, Ed Jaffe wrote: >Empirical testing shows R0 is returned with a clean 64-bit address, even >for an old 24-bit program: I believe it, but do we want to rely on undocumented behavior? Perhaps an RCF is in order? -- Tom Marchant
Re: Csect - Dsect Question
Ol' friend, your test is incomplete. That LMG doesn't go far enough. Bob On 4/1/2016 3:36 PM, Ed Jaffe wrote: On 3/31/2016 2:03 PM, Tom Marchant wrote: ITYM R0. Indeed! And the manual doesn't specify that the address returned is a clean 64-bit address except if it is AMODE 64. So I'd suggest replacing the NILH with LLGTR R0,R0 Empirical testing shows R0 is returned with a clean 64-bit address, even for an old 24-bit program: TCB#6 RB#1 - z/XDC TPUT INTERFACE XDC ===> _ _0E957100 0s (A.S.EDJX2) --- EJESSUBS.EJESSUBS+2100, EJESSUBS+2100, XPRIVATE+1A57100 _ +2100 *** _ +2100* Empirical Test for My ASSEMBLER-LIST Friends* _ +2100 *** _ +2100 EBEF C060 0004 LMG R14,R15,=4FD'-1' Put garbage in R14-R1 _ +2106 LOAD EPLOC==CL8'IEFBR14' Load a module _ +2106 4100 C0C0 + LA0,=CL8'IEFBR14' LOAD PARAMETER INTO REG 0 _ +210A 1B11 + SR1,1 SHOW NO DCB PRESENT _ +210C 0A08 + SVC 8 _ +210E > DCH'0' Force an abend _ +210E > +0 *..* XDC ===> L RWREGS _ RW0 _00EC5000 _0001 _7F363434 _ _ RW4 _7F33A71E _0E9747CE _0041 _F000 _ RW8 _7F466D00 _7F33C1A4 _7F33A000 _0E955000 _ RW12 _0E98E7B0 _7F466E00 _ 0001_
Re: Csect - Dsect Question
On 4/1/2016 12:48 PM, Bob Rutledge wrote: Ol' friend, your test is incomplete. That LMG doesn't go far enough. Sorry. Pasted from the wrong test... :-[ TCB#6 RB#1 - z/XDC TPUT INTERFACE XDC ===> _ _0E957100 0s (A.S.EDJX2) --- EJESSUBS.EJESSUBS+2100, EJESSUBS+2100, XPRIVATE+1A57100 _ +2100 *** _ +2100* Empirical Test for My ASSEMBLER-LIST Friends* _ +2100 *** _ +2100 EBE1 C060 0004 LMG R14,R1,=4FD'-1' Put garbage in R14-R1 _ +2106 LOAD EPLOC==CL8'IEFBR14' Load a module _ +2106 4100 C0C0 + LA0,=CL8'IEFBR14' LOAD PARAMETER INTO REG 0 _ +210A 1B11 + SR1,1 SHOW NO DCB PRESENT _ +210C 0A08 + SVC 8 _ +210E > DCH'0' Force an abend _ +210E > +0 *..* XDC ===> L RWREGS _ RW0 _00EC5000 _0001 _7F363434 _ _ RW4 _7F33A71E _0E9747CE _0041 _F000 _ RW8 _7F466D00 _7F33C1A4 _7F33A000 _0E955000 _ RW12 _0E98E7B0 _7F466E00 _ 0001_ -- Edward E Jaffe Phoenix Software International, Inc 831 Parkview Drive North El Segundo, CA 90245 http://www.phoenixsoftware.com/
Re: Csect - Dsect Question
>Empirical testing shows R0 is returned with a clean 64-bit address, even >for an old 24-bit program. Of course there are other forms of LOAD (those that use SVC 7A). So a test of SVC 8 is not even a complete empirical test. And of course none of the IBM-Main readers would ever rely on an empirical test. The intent of the doc is to show that the 64-bit GR0 can be used as a BASSM target in all cases. I'll get it updated to make it clear that bits 0-31 are set to 0's for the AMODE 24 and AMODE 31 cases, and are not necessarily 0 for the AMODE 64 case when LOAD with ADDR64 was used. And if RMODE 64 is ever supported for normal LOAD, then the AMODE 64 case would change too. Peter Relson z/OS Core Technology Design
Re: Csect - Dsect Question
On 4/2/2016 6:51 AM, Peter Relson wrote: And of course none of the IBM-Main readers would ever rely on an empirical test. Haha! Of course not! ;-) (But, ASSEMBLER-LIST readers might ijs... :-) ) The intent of the doc is to show that the 64-bit GR0 can be used as a BASSM target in all cases. I'll get it updated to make it clear that bits 0-31 are set to 0's for the AMODE 24 and AMODE 31 cases, and are not necessarily 0 for the AMODE 64 case when LOAD with ADDR64 was used. And if RMODE 64 is ever supported for normal LOAD, then the AMODE 64 case would change too. Thanks, Peter! That's a big help! Otherwise, people will be putting LLGTRs into their code and then removing them (or forgetting to) if/when RMODE(64) becomes reality... With all of this in mind, I suppose I should amend my example to read: LOAD EPLOC==CL8'MSGTABLE' Load external message table TMLL R0,X'0001' RMODE(64) ? JO*+8 ---\ Branch if yes NIHL R0,X'7FFF' | Turn of 31-bit indicator NILL R0,X'FFFE'
Re: Csect - Dsect Question
Guys, This is great for me. Have don't any 64bit Assembler work and want to..so I am a sponge here... Scott On Saturday, April 2, 2016, Ed Jaffe wrote: > On 4/2/2016 6:51 AM, Peter Relson wrote: > >> And of course none of the IBM-Main readers would ever rely on an >> empirical test. >> > > Haha! Of course not! ;-) > > (But, ASSEMBLER-LIST readers might ijs... :-) ) > > The intent of the doc is to show that the 64-bit GR0 can be used as a >> BASSM target in all cases. I'll get it updated to make it clear that bits >> 0-31 are set to 0's for the AMODE 24 and AMODE 31 cases, and are not >> necessarily 0 for the AMODE 64 case when LOAD with ADDR64 was used. And if >> RMODE 64 is ever supported for normal LOAD, then the AMODE 64 case would >> change too. >> > > Thanks, Peter! That's a big help! > > Otherwise, people will be putting LLGTRs into their code and then removing > them (or forgetting to) if/when RMODE(64) becomes reality... > > With all of this in mind, I suppose I should amend my example to read: > > LOAD EPLOC==CL8'MSGTABLE' Load external message table > TMLL R0,X'0001' RMODE(64) ? > JO*+8 ---\ Branch if yes > NIHL R0,X'7FFF' | Turn of 31-bit indicator > NILL R0,X'FFFE' STG R0,MsgTable Save clean table address > > -- > Edward E Jaffe > Phoenix Software International, Inc > 831 Parkview Drive North > El Segundo, CA 90245 > http://www.phoenixsoftware.com/ >
Re: Csect - Dsect Question
NILH, not NIHL! Sheesh, assembler is hard :-) Too bad it's been obviated at this point, I was going to suggest NILF R0,x'7ffe'. sas On 4/2/2016 12:05, Ed Jaffe wrote: On 4/2/2016 6:51 AM, Peter Relson wrote: And of course none of the IBM-Main readers would ever rely on an empirical test. Haha! Of course not! ;-) (But, ASSEMBLER-LIST readers might ijs... :-) ) The intent of the doc is to show that the 64-bit GR0 can be used as a BASSM target in all cases. I'll get it updated to make it clear that bits 0-31 are set to 0's for the AMODE 24 and AMODE 31 cases, and are not necessarily 0 for the AMODE 64 case when LOAD with ADDR64 was used. And if RMODE 64 is ever supported for normal LOAD, then the AMODE 64 case would change too. Thanks, Peter! That's a big help! Otherwise, people will be putting LLGTRs into their code and then removing them (or forgetting to) if/when RMODE(64) becomes reality... With all of this in mind, I suppose I should amend my example to read: LOAD EPLOC==CL8'MSGTABLE' Load external message table TMLL R0,X'0001' RMODE(64) ? JO*+8 ---\ Branch if yes NIHL R0,X'7FFF' | Turn of 31-bit indicator NILL R0,X'FFFE'
Re: Csect - Dsect Question
On 4/2/2016 9:22 AM, Steve Smith wrote: NILH, not NIHL! LOL. I should have quit while I was ahead. This one has been desk-checked. (I even corrected the misspelling of the word "off"): LOAD EPLOC==CL8'MSGTABLE' Load external message table TMLL R0,X'0001' RMODE(64) ? JO*+8 ---\ Branch if yes NILH R0,X'7FFF' | Turn off 31-bit indicator NILL R0,X'FFFE'
