Re: BSAM READ AMODE 31 RMODE 31 does I/O area have to be fixed ?
Hi This post is somewhat related to the original Post I was just looking at the parameters for DCBE and one of them FIXED=USER with this DCBE option you assert that data areas remain fixed from the time the read or write is issued through the completion of WAIT or CHECK So first I was unaware of this Second does this refer to the I/O areas Or the control block DCB, DCBE DECB as well I was able to get my READ working when I chose sp=233 for the I/O area which is it not pageable and fixed Thanks > On Mar 22, 2023, at 8:09 AM, reichman...@gmail.com wrote: > > This is the code all I did was change the ioarea from sp=0 to sp=233 > I am running under TESTAUTH and all my routines start by establishing an > estate > > After the balr from the read (using sp = 0) that's were it went to the estae > > Checked R1-> SDWAPTR at +5 for 12 bits was 800 r15 starting at +18 from > sdwa was a 4 > > thanks >L R0,=F'31996' > * > MVCSUP_LIST(SUP_LEN),SUP_ONMOVE CONSTANT > * > MODESET MF=(E,SUP_LIST) STORAGE KEY ZERO > * > SPKA 0 > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=(31,64),SP=233 > <-- > *LAR7,IOAREA > *STR7,IOPTR > * > STR7,IOBUFF > L R7,IOBUFF > MVC OPEN_LST(OPEN_LEN),OPEN_CON > OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 > * > *MVC ADATADECB(IREADLEN),IREAD > TMDCBOFLGS,DCBOFOPN > BZEXITDEBG > USING ASMADATA,R7 > PROCADATA DS 0H > READ ADATADECB,SF,(R6),(R7),'S',MF=E > * > CHECK ADATADECB > * > LAR7,4(,R7) > CHSI ADATA_RECTYPE,ADATA_RECESD External sym dic. > BNE CKSRC > > > -Original Message- > From: IBM Mainframe Discussion List On Behalf Of > Binyamin Dissen > Sent: Wednesday, March 22, 2023 5:24 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: BSAM READ AMODE 31 RMODE 31 > > Did you check the dump as I suggested? > > If so, post your new code. > > On Wed, 22 Mar 2023 03:54:37 -0400 Joseph Reichman > wrote: > > :>This was when I used R7 R7 had the address of the IOAREA it abended 800 RC > 4 when I got the storage from sp 233 it worked :> :>This is application code > subpool zero shouldn't give me a problem :> :>> On Mar 22, 2023, at 2:48 AM, > Binyamin Dissen wrote: > :>> > :>> ?Because when you supply IOBUFF as the address of the buffer it will > overlay :>> storage following that label. That includes register save areas, > code, etc. > :>> Any kind of abend may occur. > :>> > :>> Look at your dump very carefully. See if the entire working storage is > as :>> expected. > :>> > :>> Look at the trace table before the abend and t6he BEA. How did you get > there? > :>> > :>> > :>> On Tue, 21 Mar 2023 21:26:20 -0400 Joseph Reichman > :>> wrote: > :>> > :>> :>This totally crazy I abended 800 RC 4 :>> :>Looked at the abend codes > says system EXCPVSR was trying to page fix storage :>> :> :>> :>I then got > sp 233 says it's fixed and also key 0 :>> :>Worked :>> :> :>> :>I don't get > it there is no reason for the IOAREA to be sp 233 subpool 0 should be good > enough :>> :> :>> :> :>> :> :>> :>> On Mar 21, 2023, at 11:43 AM, Seymour J > Metz wr
Re: BSAM READ AMODE 31 RMODE 31
This is the code all I did was change the ioarea from sp=0 to sp=233 I am running under TESTAUTH and all my routines start by establishing an estate After the balr from the read (using sp = 0) that's were it went to the estae Checked R1-> SDWAPTR at +5 for 12 bits was 800 r15 starting at +18 from sdwa was a 4 thanks L R0,=F'31996' * MVCSUP_LIST(SUP_LEN),SUP_ONMOVE CONSTANT * MODESET MF=(E,SUP_LIST) STORAGE KEY ZERO * SPKA 0 STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=(31,64),SP=233 <-- *LAR7,IOAREA *STR7,IOPTR * STR7,IOBUFF L R7,IOBUFF MVC OPEN_LST(OPEN_LEN),OPEN_CON OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 * *MVC ADATADECB(IREADLEN),IREAD TMDCBOFLGS,DCBOFOPN BZEXITDEBG USING ASMADATA,R7 PROCADATA DS 0H READ ADATADECB,SF,(R6),(R7),'S',MF=E * CHECK ADATADECB * LAR7,4(,R7) CHSI ADATA_RECTYPE,ADATA_RECESD External sym dic. BNE CKSRC -Original Message- From: IBM Mainframe Discussion List On Behalf Of Binyamin Dissen Sent: Wednesday, March 22, 2023 5:24 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: BSAM READ AMODE 31 RMODE 31 Did you check the dump as I suggested? If so, post your new code. On Wed, 22 Mar 2023 03:54:37 -0400 Joseph Reichman wrote: :>This was when I used R7 R7 had the address of the IOAREA it abended 800 RC 4 when I got the storage from sp 233 it worked :> :>This is application code subpool zero shouldn't give me a problem :> :>> On Mar 22, 2023, at 2:48 AM, Binyamin Dissen wrote: :>> :>> ?Because when you supply IOBUFF as the address of the buffer it will overlay :>> storage following that label. That includes register save areas, code, etc. :>> Any kind of abend may occur. :>> :>> Look at your dump very carefully. See if the entire working storage is as :>> expected. :>> :>> Look at the trace table before the abend and t6he BEA. How did you get there? :>> :>> :>> On Tue, 21 Mar 2023 21:26:20 -0400 Joseph Reichman :>> wrote: :>> :>> :>This totally crazy I abended 800 RC 4 :>> :>Looked at the abend codes says system EXCPVSR was trying to page fix storage :>> :> :>> :>I then got sp 233 says it's fixed and also key 0 :>> :>Worked :>> :> :>> :>I don't get it there is no reason for the IOAREA to be sp 233 subpool 0 should be good enough :>> :> :>> :> :>> :> :>> :>> On Mar 21, 2023, at 11:43 AM, Seymour J Metz wrote: :>> :>> :>> :>> ?You have to give READ the address of the buffer, not the address of a pointer to it. :>> :>> :>> :>> For 64-bit there is a different DECB format; I believe that's SF64. :>> :>> :>> :>> :>> :>> From: IBM Mainframe Discussion List on behalf of Joseph Reichman :>> :>> Sent: Tuesday, March 21, 2023 11:22 AM :>> :>> To: IBM-MAIN@LISTSERV.UA.EDU :>> :>> Subject: Re: BSAM READ AMODE 31 RMODE 31 :>> :>> :>> :>> Still have problems :>> :>> :>> :>> With the read :>> :>> :>> :>> I tried last night and it worked let me explain :>> :>> :>> :>> The address of the IO area I store in a full word :>> :>> IOBUFF :>> :>> :>> :>> When I did LA R7,IOBUFF :&
Re: BSAM READ AMODE 31 RMODE 31
Did you check the dump as I suggested? If so, post your new code. On Wed, 22 Mar 2023 03:54:37 -0400 Joseph Reichman wrote: :>This was when I used R7 R7 had the address of the IOAREA it abended 800 RC 4 when I got the storage from sp 233 it worked :> :>This is application code subpool zero shouldnt give me a problem :> :>> On Mar 22, 2023, at 2:48 AM, Binyamin Dissen wrote: :>> :>> ?Because when you supply IOBUFF as the address of the buffer it will overlay :>> storage following that label. That includes register save areas, code, etc. :>> Any kind of abend may occur. :>> :>> Look at your dump very carefully. See if the entire working storage is as :>> expected. :>> :>> Look at the trace table before the abend and t6he BEA. How did you get there? :>> :>> :>> On Tue, 21 Mar 2023 21:26:20 -0400 Joseph Reichman :>> wrote: :>> :>> :>This totally crazy I abended 800 RC 4 :>> :>Looked at the abend codes says system EXCPVSR was trying to page fix storage :>> :> :>> :>I then got sp 233 says its fixed and also key 0 :>> :>Worked :>> :> :>> :>I dont get it there is no reason for the IOAREA to be sp 233 subpool 0 should be good enough :>> :> :>> :> :>> :> :>> :>> On Mar 21, 2023, at 11:43 AM, Seymour J Metz wrote: :>> :>> :>> :>> ?You have to give READ the address of the buffer, not the address of a pointer to it. :>> :>> :>> :>> For 64-bit there is a different DECB format; I believe that's SF64. :>> :>> :>> :>> :>> :>> From: IBM Mainframe Discussion List on behalf of Joseph Reichman :>> :>> Sent: Tuesday, March 21, 2023 11:22 AM :>> :>> To: IBM-MAIN@LISTSERV.UA.EDU :>> :>> Subject: Re: BSAM READ AMODE 31 RMODE 31 :>> :>> :>> :>> Still have problems :>> :>> :>> :>> With the read :>> :>> :>> :>> I tried last night and it worked let me explain :>> :>> :>> :>> The address of the IO area I store in a full word :>> :>> IOBUFF :>> :>> :>> :>> When I did LA R7,IOBUFF :>> :>> And then did :>> :>> :>> :>> READ ADATADECB,SF,(R6),(R7),S,MF=E :>> :>> :>> :>> It worked today morning didnt change code had problems :>> :>> :>> :>> When I did L R7,IOBUFF it didnt :>> :>> :>> :>> I know both Binyamin and Charles Mills in their example had Load but it doesnt make sense :>> :>> :>> :>> At +12 is the address of a pointer to the IOAREA if that were the address it self :>> :>> :>> :>> What if you had a 64 bit pointer (which bsam supoorts) so it had to be a pointer to where the pointer ( in 64 bit that would 8 bytes ) :>> :>> Thanks :>> :>> :>> :>> :>> :>>> On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: :>> :>>> :>> :>>> ?IMHO it's clearer to just use :>> :>>> :>> :>>> READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E :>> :>>> CHECK PDSDECB :>> :>>> :>> :>>> :>> :>>> -- :>> :>>> Shmuel (Seymour J.) Metz :>> :>>> http://mason.gmu.edu/~smetz3 :>> :>>> :>> :>>> :>> :>>> From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Joseph Reichman [reichman...@gmail.com] :>> :>>> Sent: Monday, March 20, 2023 9:20 AM :>> :>>> To: IBM-MAIN@LISTSERV.UA.EDU :>> :>>> Subject: BSAM READ AMODE 31 RMODE 31 :>> :>>> :>> :>>> Hi :>> :>>> :>> :>>> :>> :>>> :>> :>>> I know Charles Mills answered my question nearly 4 weeks ago but I would :>> :>>> just like to corroborate the I/O areas for the read can be 31 bit addressing :>> :>>> only the dcb and decb need to be below the line :>> :>>> :>> :>>> I am running AMODE 31 RMODE 31 :>> :>>> :>> :>>> :>> :>>> :>> :>>> The DCB is below the line as well as the DECB :>> :>>> :>> :>>> :>> :>>> :>> :>>> My read returns ok :>> :>>> :>> :>>> :>> :>>> :>> :>>> After I set a beak point after the CHECK I get
Re: BSAM READ AMODE 31 RMODE 31
This was when I used R7 R7 had the address of the IOAREA it abended 800 RC 4 when I got the storage from sp 233 it worked This is application code subpool zero shouldn’t give me a problem > On Mar 22, 2023, at 2:48 AM, Binyamin Dissen > wrote: > > Because when you supply IOBUFF as the address of the buffer it will overlay > storage following that label. That includes register save areas, code, etc. > Any kind of abend may occur. > > Look at your dump very carefully. See if the entire working storage is as > expected. > > Look at the trace table before the abend and t6he BEA. How did you get there? > > > On Tue, 21 Mar 2023 21:26:20 -0400 Joseph Reichman > wrote: > > :>This totally crazy I abended 800 RC 4 > :>Looked at the abend codes says system EXCPVSR was trying to page fix > storage > :> > :>I then got sp 233 says it’s fixed and also key 0 > :>Worked > :> > :>I don’t get it there is no reason for the IOAREA to be sp 233 subpool 0 > should be good enough > :> > :> > :> > :>> On Mar 21, 2023, at 11:43 AM, Seymour J Metz wrote: > :>> > :>> ?You have to give READ the address of the buffer, not the address of a > pointer to it. > :>> > :>> For 64-bit there is a different DECB format; I believe that's SF64. > :>> > :>> ________ > :>> From: IBM Mainframe Discussion List on behalf > of Joseph Reichman > :>> Sent: Tuesday, March 21, 2023 11:22 AM > :>> To: IBM-MAIN@LISTSERV.UA.EDU > :>> Subject: Re: BSAM READ AMODE 31 RMODE 31 > :>> > :>> Still have problems > :>> > :>> With the read > :>> > :>> I tried last night and it worked let me explain > :>> > :>> The address of the IO area I store in a full word > :>> IOBUFF > :>> > :>> When I did LA R7,IOBUFF > :>> And then did > :>> > :>> READ ADATADECB,SF,(R6),(R7),’S’,MF=E > :>> > :>> It worked today morning didn’t change code had problems > :>> > :>> When I did L R7,IOBUFF it didn’t > :>> > :>> I know both Binyamin and Charles Mills in their example had Load but it > doesn’t make sense > :>> > :>> At +12 is the address of a pointer to the IOAREA if that were the address > it self > :>> > :>> What if you had a 64 bit pointer (which bsam supoorts) so it had to be a > pointer to where the pointer ( in 64 bit that would 8 bytes ) > :>> Thanks > :>> > :>> > :>>> On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: > :>>> > :>>> ?IMHO it's clearer to just use > :>>> > :>>> READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E > :>>> CHECK PDSDECB > :>>> > :>>> > :>>> -- > :>>> Shmuel (Seymour J.) Metz > :>>> http://mason.gmu.edu/~smetz3 > :>>> > :>>> > :>>> From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf > of Joseph Reichman [reichman...@gmail.com] > :>>> Sent: Monday, March 20, 2023 9:20 AM > :>>> To: IBM-MAIN@LISTSERV.UA.EDU > :>>> Subject: BSAM READ AMODE 31 RMODE 31 > :>>> > :>>> Hi > :>>> > :>>> > :>>> > :>>> I know Charles Mills answered my question nearly 4 weeks ago but I would > :>>> just like to corroborate the I/O areas for the read can be 31 bit > addressing > :>>> only the dcb and decb need to be below the line > :>>> > :>>> I am running AMODE 31 RMODE 31 > :>>> > :>>> > :>>> > :>>> The DCB is below the line as well as the DECB > :>>> > :>>> > :>>> > :>>> My read returns ok > :>>> > :>>> > :>>> > :>>> After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP > :>>> > :>>> > :>>> > :>>> This the example that Charles sent to me does nt say anything about the > :>>> IOAREAS or pointers to the IOAREAS > :>>> > :>>> > :>>> > :>>> LA Rn,PDSDCB > :>>> > :>>> ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB > :>>> > :>>> L Rn,address of READ buffer > :>>> > :>>> READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB > :>>> > :>>> > :>>
Re: BSAM READ AMODE 31 RMODE 31
Because when you supply IOBUFF as the address of the buffer it will overlay storage following that label. That includes register save areas, code, etc. Any kind of abend may occur. Look at your dump very carefully. See if the entire working storage is as expected. Look at the trace table before the abend and t6he BEA. How did you get there? On Tue, 21 Mar 2023 21:26:20 -0400 Joseph Reichman wrote: :>This totally crazy I abended 800 RC 4 :>Looked at the abend codes says system EXCPVSR was trying to page fix storage :> :>I then got sp 233 says its fixed and also key 0 :>Worked :> :>I dont get it there is no reason for the IOAREA to be sp 233 subpool 0 should be good enough :> :> :> :>> On Mar 21, 2023, at 11:43 AM, Seymour J Metz wrote: :>> :>> ?You have to give READ the address of the buffer, not the address of a pointer to it. :>> :>> For 64-bit there is a different DECB format; I believe that's SF64. :>> :>> :>> From: IBM Mainframe Discussion List on behalf of Joseph Reichman :>> Sent: Tuesday, March 21, 2023 11:22 AM :>> To: IBM-MAIN@LISTSERV.UA.EDU :>> Subject: Re: BSAM READ AMODE 31 RMODE 31 :>> :>> Still have problems :>> :>> With the read :>> :>> I tried last night and it worked let me explain :>> :>> The address of the IO area I store in a full word :>> IOBUFF :>> :>> When I did LA R7,IOBUFF :>> And then did :>> :>> READ ADATADECB,SF,(R6),(R7),S,MF=E :>> :>> It worked today morning didnt change code had problems :>> :>> When I did L R7,IOBUFF it didnt :>> :>> I know both Binyamin and Charles Mills in their example had Load but it doesnt make sense :>> :>> At +12 is the address of a pointer to the IOAREA if that were the address it self :>> :>> What if you had a 64 bit pointer (which bsam supoorts) so it had to be a pointer to where the pointer ( in 64 bit that would 8 bytes ) :>> Thanks :>> :>> :>>> On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: :>>> :>>> ?IMHO it's clearer to just use :>>> :>>> READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E :>>> CHECK PDSDECB :>>> :>>> :>>> -- :>>> Shmuel (Seymour J.) Metz :>>> http://mason.gmu.edu/~smetz3 :>>> :>>> :>>> From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Joseph Reichman [reichman...@gmail.com] :>>> Sent: Monday, March 20, 2023 9:20 AM :>>> To: IBM-MAIN@LISTSERV.UA.EDU :>>> Subject: BSAM READ AMODE 31 RMODE 31 :>>> :>>> Hi :>>> :>>> :>>> :>>> I know Charles Mills answered my question nearly 4 weeks ago but I would :>>> just like to corroborate the I/O areas for the read can be 31 bit addressing :>>> only the dcb and decb need to be below the line :>>> :>>> I am running AMODE 31 RMODE 31 :>>> :>>> :>>> :>>> The DCB is below the line as well as the DECB :>>> :>>> :>>> :>>> My read returns ok :>>> :>>> :>>> :>>> After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP :>>> :>>> :>>> :>>> This the example that Charles sent to me does nt say anything about the :>>> IOAREAS or pointers to the IOAREAS :>>> :>>> :>>> :>>> LA Rn,PDSDCB :>>> :>>> ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB :>>> :>>> L Rn,address of READ buffer :>>> :>>> READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB :>>> :>>> :>>> :>>> :>>> :>>> :>>> :>>> :>>> :>>> This is my read :>>> :>>> :>>> :>>> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E :>>> :>>> :>>> :>>> The IOBUFF is a full word pointer to the storage I obtain :>>> :>>> :>>> :>>> :>>> :>>> Here is my code :>>> :>>> :>>> :>>> LAR0,IODSECTLEN :>>> :>>> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 :>>> :>>> LRR9,R6 :>>> :>>> USING IHADCB,R9 :>>> :>>> USING IODSECT,R6 :>>> :>>> * :>>> :>>> * :>>> :>>&g
Re: BSAM READ AMODE 31 RMODE 31
This totally crazy I abended 800 RC 4 Looked at the abend codes says system EXCPVSR was trying to page fix storage I then got sp 233 says it’s fixed and also key 0 Worked I don’t get it there is no reason for the IOAREA to be sp 233 subpool 0 should be good enough > On Mar 21, 2023, at 11:43 AM, Seymour J Metz wrote: > > You have to give READ the address of the buffer, not the address of a > pointer to it. > > For 64-bit there is a different DECB format; I believe that's SF64. > > > From: IBM Mainframe Discussion List on behalf of > Joseph Reichman > Sent: Tuesday, March 21, 2023 11:22 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: BSAM READ AMODE 31 RMODE 31 > > Still have problems > > With the read > > I tried last night and it worked let me explain > > The address of the IO area I store in a full word > IOBUFF > > When I did LA R7,IOBUFF > And then did > > READ ADATADECB,SF,(R6),(R7),’S’,MF=E > > It worked today morning didn’t change code had problems > > When I did L R7,IOBUFF it didn’t > > I know both Binyamin and Charles Mills in their example had Load but it > doesn’t make sense > > At +12 is the address of a pointer to the IOAREA if that were the address it > self > > What if you had a 64 bit pointer (which bsam supoorts) so it had to be a > pointer to where the pointer ( in 64 bit that would 8 bytes ) > Thanks > > >> On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: >> >> IMHO it's clearer to just use >> >> READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E >> CHECK PDSDECB >> >> >> -- >> Shmuel (Seymour J.) Metz >> http://mason.gmu.edu/~smetz3 >> >> >> From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of >> Joseph Reichman [reichman...@gmail.com] >> Sent: Monday, March 20, 2023 9:20 AM >> To: IBM-MAIN@LISTSERV.UA.EDU >> Subject: BSAM READ AMODE 31 RMODE 31 >> >> Hi >> >> >> >> I know Charles Mills answered my question nearly 4 weeks ago but I would >> just like to corroborate the I/O areas for the read can be 31 bit addressing >> only the dcb and decb need to be below the line >> >> I am running AMODE 31 RMODE 31 >> >> >> >> The DCB is below the line as well as the DECB >> >> >> >> My read returns ok >> >> >> >> After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP >> >> >> >> This the example that Charles sent to me does nt say anything about the >> IOAREAS or pointers to the IOAREAS >> >> >> >> LA Rn,PDSDCB >> >> ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB >> >> L Rn,address of READ buffer >> >> READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB >> >> >> >> >> >> >> >> >> >> This is my read >> >> >> >> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E >> >> >> >> The IOBUFF is a full word pointer to the storage I obtain >> >> >> >> >> >> Here is my code >> >> >> >> LAR0,IODSECTLEN >> >> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 >> >> LRR9,R6 >> >> USING IHADCB,R9 >> >> USING IODSECT,R6 >> >> * >> >> * >> >> LAR7,SYSADATA >> >> MVC 0(IODSECTLEN,R6),0(R7) >> >> * >> >> LAR7,SYSDCBE-SYSADATA(,R6) >> >> STR7,DCBDCBE >> >> LAR7,EX24LST-SYSADATA(,R6) >> >> STCM R7,B'0111',DCBEXLSA >> >> LAR7,ABND24-SYSADATA(,R6) >> >> STCM R7,B'0111',ABND24ADR-SYSADATA(R6) >> >> LAR7,ABENDRTN >> >> STR7,ABEND31-SYSADATA(,R6) >> >> * >> >> T L R0,=F'31996' >> >> * >> >> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 >> >> *LAR7,IOAREA >> >> *STR7,IOPTR >> >> * >> >> STR7,IOBUFF >> >> MVC OPEN_LST(OPEN_LEN),OPEN_CON >> >> OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 >> >> * >> >> TMDCBOFLGS,DCBOFOPN >> >
Re: BSAM READ AMODE 31 RMODE 31
All I want to say then is the comment next to the ST Rx,12(R1) says store 64 bit pointer not saying you are not right Thank you > On Mar 21, 2023, at 11:43 AM, Seymour J Metz wrote: > > You have to give READ the address of the buffer, not the address of a > pointer to it. > > For 64-bit there is a different DECB format; I believe that's SF64. > > > From: IBM Mainframe Discussion List on behalf of > Joseph Reichman > Sent: Tuesday, March 21, 2023 11:22 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: BSAM READ AMODE 31 RMODE 31 > > Still have problems > > With the read > > I tried last night and it worked let me explain > > The address of the IO area I store in a full word > IOBUFF > > When I did LA R7,IOBUFF > And then did > > READ ADATADECB,SF,(R6),(R7),’S’,MF=E > > It worked today morning didn’t change code had problems > > When I did L R7,IOBUFF it didn’t > > I know both Binyamin and Charles Mills in their example had Load but it > doesn’t make sense > > At +12 is the address of a pointer to the IOAREA if that were the address it > self > > What if you had a 64 bit pointer (which bsam supoorts) so it had to be a > pointer to where the pointer ( in 64 bit that would 8 bytes ) > Thanks > > >> On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: >> >> IMHO it's clearer to just use >> >> READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E >> CHECK PDSDECB >> >> >> -- >> Shmuel (Seymour J.) Metz >> http://mason.gmu.edu/~smetz3 >> >> >> From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of >> Joseph Reichman [reichman...@gmail.com] >> Sent: Monday, March 20, 2023 9:20 AM >> To: IBM-MAIN@LISTSERV.UA.EDU >> Subject: BSAM READ AMODE 31 RMODE 31 >> >> Hi >> >> >> >> I know Charles Mills answered my question nearly 4 weeks ago but I would >> just like to corroborate the I/O areas for the read can be 31 bit addressing >> only the dcb and decb need to be below the line >> >> I am running AMODE 31 RMODE 31 >> >> >> >> The DCB is below the line as well as the DECB >> >> >> >> My read returns ok >> >> >> >> After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP >> >> >> >> This the example that Charles sent to me does nt say anything about the >> IOAREAS or pointers to the IOAREAS >> >> >> >> LA Rn,PDSDCB >> >> ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB >> >> L Rn,address of READ buffer >> >> READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB >> >> >> >> >> >> >> >> >> >> This is my read >> >> >> >> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E >> >> >> >> The IOBUFF is a full word pointer to the storage I obtain >> >> >> >> >> >> Here is my code >> >> >> >> LAR0,IODSECTLEN >> >> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 >> >> LRR9,R6 >> >> USING IHADCB,R9 >> >> USING IODSECT,R6 >> >> * >> >> * >> >> LAR7,SYSADATA >> >> MVC 0(IODSECTLEN,R6),0(R7) >> >> * >> >> LAR7,SYSDCBE-SYSADATA(,R6) >> >> STR7,DCBDCBE >> >> LAR7,EX24LST-SYSADATA(,R6) >> >> STCM R7,B'0111',DCBEXLSA >> >> LAR7,ABND24-SYSADATA(,R6) >> >> STCM R7,B'0111',ABND24ADR-SYSADATA(R6) >> >> LAR7,ABENDRTN >> >> STR7,ABEND31-SYSADATA(,R6) >> >> * >> >> T L R0,=F'31996' >> >> * >> >> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 >> >> *LAR7,IOAREA >> >> *STR7,IOPTR >> >> * >> >> STR7,IOBUFF >> >> MVC OPEN_LST(OPEN_LEN),OPEN_CON >> >> OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 >> >> * >> >> TMDCBOFLGS,DCBOFOPN >> >> BZEXITDEBG >&
Re: BSAM READ AMODE 31 RMODE 31
You have to give READ the address of the buffer, not the address of a pointer to it. For 64-bit there is a different DECB format; I believe that's SF64. From: IBM Mainframe Discussion List on behalf of Joseph Reichman Sent: Tuesday, March 21, 2023 11:22 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: BSAM READ AMODE 31 RMODE 31 Still have problems With the read I tried last night and it worked let me explain The address of the IO area I store in a full word IOBUFF When I did LA R7,IOBUFF And then did READ ADATADECB,SF,(R6),(R7),’S’,MF=E It worked today morning didn’t change code had problems When I did L R7,IOBUFF it didn’t I know both Binyamin and Charles Mills in their example had Load but it doesn’t make sense At +12 is the address of a pointer to the IOAREA if that were the address it self What if you had a 64 bit pointer (which bsam supoorts) so it had to be a pointer to where the pointer ( in 64 bit that would 8 bytes ) Thanks > On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: > > IMHO it's clearer to just use > > READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E > CHECK PDSDECB > > > -- > Shmuel (Seymour J.) Metz > http://mason.gmu.edu/~smetz3 > > > From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of > Joseph Reichman [reichman...@gmail.com] > Sent: Monday, March 20, 2023 9:20 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: BSAM READ AMODE 31 RMODE 31 > > Hi > > > > I know Charles Mills answered my question nearly 4 weeks ago but I would > just like to corroborate the I/O areas for the read can be 31 bit addressing > only the dcb and decb need to be below the line > > I am running AMODE 31 RMODE 31 > > > > The DCB is below the line as well as the DECB > > > > My read returns ok > > > > After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP > > > > This the example that Charles sent to me does nt say anything about the > IOAREAS or pointers to the IOAREAS > > > > LA Rn,PDSDCB > > ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB > > L Rn,address of READ buffer > > READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB > > > > > > > > > > This is my read > > > > READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > > > > The IOBUFF is a full word pointer to the storage I obtain > > > > > > Here is my code > > > > LAR0,IODSECTLEN > > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 > > LRR9,R6 > > USING IHADCB,R9 > > USING IODSECT,R6 > > * > > * > > LAR7,SYSADATA > > MVC 0(IODSECTLEN,R6),0(R7) > > * > > LAR7,SYSDCBE-SYSADATA(,R6) > > STR7,DCBDCBE > > LAR7,EX24LST-SYSADATA(,R6) > > STCM R7,B'0111',DCBEXLSA > > LAR7,ABND24-SYSADATA(,R6) > > STCM R7,B'0111',ABND24ADR-SYSADATA(R6) > > LAR7,ABENDRTN > > STR7,ABEND31-SYSADATA(,R6) > > * > > T L R0,=F'31996' > > * > > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 > > *LAR7,IOAREA > > *STR7,IOPTR > > * > > STR7,IOBUFF > > MVC OPEN_LST(OPEN_LEN),OPEN_CON > > OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 > > * > > TMDCBOFLGS,DCBOFOPN > > BZEXITDEBG > > USING ASMADATA,R7 > > L R10,IOBUFF > > PROCADATA DS 0H > > READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > > * > > CHECK ADATADECB > > * > > OPEN_CON OPEN (*-*,INPUT),MF=L,MODE=31 > > SYSADATA DCB DDNAME=SYSADATA,RECFM=VB,MACRF=RP,DSORG=PS,DCBE=SYSDCBE,EXX > > LST=EX24LST > > > > IHADECB DSECT=NO > > DECBLEN EQU *-DECB > > SYSDCBE DCBE EODAD=FIN,GETSIZE=YES,LOC=ANY,RMODE31=BUFF,SYNAD=SYNAD,VEX > > RSION=1 > > EX24LSTDS 0F > > ADATAEXLST DC AL1(EXLDCBAB+X'80') > > ABND24ADR DS AL3 > > ABND24 DS 0H > > L R15,ABEND31-SYSADATA(,R15) > > BSM 0,R15 > > ABEND31DS XL4 > > STROGE24 EQU *-EX24LST > > * > > IODSECT DSECT > > ADATADCB DSCL(DCBLNGBS) > > ADATADECB DSCL(DECBLEN) > > ADATADCBE DSCL(DCBELENV1) > > DS0F > > INST31DSCL(STROGE24) > > IODSECT
Re: BSAM READ AMODE 31 RMODE 31
Forgot to mention the DECB is RMODE 24 > On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: > > IMHO it's clearer to just use > > READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E > CHECK PDSDECB > > > -- > Shmuel (Seymour J.) Metz > http://mason.gmu.edu/~smetz3 > > > From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of > Joseph Reichman [reichman...@gmail.com] > Sent: Monday, March 20, 2023 9:20 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: BSAM READ AMODE 31 RMODE 31 > > Hi > > > > I know Charles Mills answered my question nearly 4 weeks ago but I would > just like to corroborate the I/O areas for the read can be 31 bit addressing > only the dcb and decb need to be below the line > > I am running AMODE 31 RMODE 31 > > > > The DCB is below the line as well as the DECB > > > > My read returns ok > > > > After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP > > > > This the example that Charles sent to me does nt say anything about the > IOAREAS or pointers to the IOAREAS > > > > LA Rn,PDSDCB > > ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB > > L Rn,address of READ buffer > > READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB > > > > > > > > > > This is my read > > > > READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > > > > The IOBUFF is a full word pointer to the storage I obtain > > > > > > Here is my code > > > > LAR0,IODSECTLEN > > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 > > LRR9,R6 > > USING IHADCB,R9 > > USING IODSECT,R6 > > * > > * > > LAR7,SYSADATA > > MVC 0(IODSECTLEN,R6),0(R7) > > * > > LAR7,SYSDCBE-SYSADATA(,R6) > > STR7,DCBDCBE > > LAR7,EX24LST-SYSADATA(,R6) > > STCM R7,B'0111',DCBEXLSA > > LAR7,ABND24-SYSADATA(,R6) > > STCM R7,B'0111',ABND24ADR-SYSADATA(R6) > > LAR7,ABENDRTN > > STR7,ABEND31-SYSADATA(,R6) > > * > > T L R0,=F'31996' > > * > > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 > > *LAR7,IOAREA > > *STR7,IOPTR > > * > > STR7,IOBUFF > > MVC OPEN_LST(OPEN_LEN),OPEN_CON > > OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 > > * > > TMDCBOFLGS,DCBOFOPN > > BZEXITDEBG > > USING ASMADATA,R7 > > L R10,IOBUFF > > PROCADATA DS 0H > > READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > > * > > CHECK ADATADECB > > * > > OPEN_CON OPEN (*-*,INPUT),MF=L,MODE=31 > > SYSADATA DCB DDNAME=SYSADATA,RECFM=VB,MACRF=RP,DSORG=PS,DCBE=SYSDCBE,EXX > > LST=EX24LST > > > > IHADECB DSECT=NO > > DECBLEN EQU *-DECB > > SYSDCBE DCBE EODAD=FIN,GETSIZE=YES,LOC=ANY,RMODE31=BUFF,SYNAD=SYNAD,VEX > > RSION=1 > > EX24LSTDS 0F > > ADATAEXLST DC AL1(EXLDCBAB+X'80') > > ABND24ADR DS AL3 > > ABND24 DS 0H > > L R15,ABEND31-SYSADATA(,R15) > > BSM 0,R15 > > ABEND31DS XL4 > > STROGE24 EQU *-EX24LST > > * > > IODSECT DSECT > > ADATADCB DSCL(DCBLNGBS) > > ADATADECB DSCL(DECBLEN) > > ADATADCBE DSCL(DCBELENV1) > > DS0F > > INST31DSCL(STROGE24) > > IODSECTLEN EQU *-IODSECT > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: BSAM READ AMODE 31 RMODE 31
Still have problems With the read I tried last night and it worked let me explain The address of the IO area I store in a full word IOBUFF When I did LA R7,IOBUFF And then did READ ADATADECB,SF,(R6),(R7),’S’,MF=E It worked today morning didn’t change code had problems When I did L R7,IOBUFF it didn’t I know both Binyamin and Charles Mills in their example had Load but it doesn’t make sense At +12 is the address of a pointer to the IOAREA if that were the address it self What if you had a 64 bit pointer (which bsam supoorts) so it had to be a pointer to where the pointer ( in 64 bit that would 8 bytes ) Thanks > On Mar 20, 2023, at 3:46 PM, Seymour J Metz wrote: > > IMHO it's clearer to just use > > READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E > CHECK PDSDECB > > > -- > Shmuel (Seymour J.) Metz > http://mason.gmu.edu/~smetz3 > > > From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of > Joseph Reichman [reichman...@gmail.com] > Sent: Monday, March 20, 2023 9:20 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: BSAM READ AMODE 31 RMODE 31 > > Hi > > > > I know Charles Mills answered my question nearly 4 weeks ago but I would > just like to corroborate the I/O areas for the read can be 31 bit addressing > only the dcb and decb need to be below the line > > I am running AMODE 31 RMODE 31 > > > > The DCB is below the line as well as the DECB > > > > My read returns ok > > > > After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP > > > > This the example that Charles sent to me does nt say anything about the > IOAREAS or pointers to the IOAREAS > > > > LA Rn,PDSDCB > > ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB > > L Rn,address of READ buffer > > READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB > > > > > > > > > > This is my read > > > > READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > > > > The IOBUFF is a full word pointer to the storage I obtain > > > > > > Here is my code > > > > LAR0,IODSECTLEN > > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 > > LRR9,R6 > > USING IHADCB,R9 > > USING IODSECT,R6 > > * > > * > > LAR7,SYSADATA > > MVC 0(IODSECTLEN,R6),0(R7) > > * > > LAR7,SYSDCBE-SYSADATA(,R6) > > STR7,DCBDCBE > > LAR7,EX24LST-SYSADATA(,R6) > > STCM R7,B'0111',DCBEXLSA > > LAR7,ABND24-SYSADATA(,R6) > > STCM R7,B'0111',ABND24ADR-SYSADATA(R6) > > LAR7,ABENDRTN > > STR7,ABEND31-SYSADATA(,R6) > > * > > T L R0,=F'31996' > > * > > STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 > > *LAR7,IOAREA > > *STR7,IOPTR > > * > > STR7,IOBUFF > > MVC OPEN_LST(OPEN_LEN),OPEN_CON > > OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 > > * > > TMDCBOFLGS,DCBOFOPN > > BZEXITDEBG > > USING ASMADATA,R7 > > L R10,IOBUFF > > PROCADATA DS 0H > > READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > > * > > CHECK ADATADECB > > * > > OPEN_CON OPEN (*-*,INPUT),MF=L,MODE=31 > > SYSADATA DCB DDNAME=SYSADATA,RECFM=VB,MACRF=RP,DSORG=PS,DCBE=SYSDCBE,EXX > > LST=EX24LST > > > > IHADECB DSECT=NO > > DECBLEN EQU *-DECB > > SYSDCBE DCBE EODAD=FIN,GETSIZE=YES,LOC=ANY,RMODE31=BUFF,SYNAD=SYNAD,VEX > > RSION=1 > > EX24LSTDS 0F > > ADATAEXLST DC AL1(EXLDCBAB+X'80') > > ABND24ADR DS AL3 > > ABND24 DS 0H > > L R15,ABEND31-SYSADATA(,R15) > > BSM 0,R15 > > ABEND31DS XL4 > > STROGE24 EQU *-EX24LST > > * > > IODSECT DSECT > > ADATADCB DSCL(DCBLNGBS) > > ADATADECB DSCL(DECBLEN) > > ADATADCBE DSCL(DCBELENV1) > > DS0F > > INST31DSCL(STROGE24) > > IODSECTLEN EQU *-IODSECT > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN > > -- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: BSAM READ AMODE 31 RMODE 31
IMHO it's clearer to just use READ PDSDECB,SF,PDSDCB,address _of_READ_buffer,MF=E CHECK PDSDECB -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Joseph Reichman [reichman...@gmail.com] Sent: Monday, March 20, 2023 9:20 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: BSAM READ AMODE 31 RMODE 31 Hi I know Charles Mills answered my question nearly 4 weeks ago but I would just like to corroborate the I/O areas for the read can be 31 bit addressing only the dcb and decb need to be below the line I am running AMODE 31 RMODE 31 The DCB is below the line as well as the DECB My read returns ok After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP This the example that Charles sent to me does nt say anything about the IOAREAS or pointers to the IOAREAS LA Rn,PDSDCB ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB L Rn,address of READ buffer READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB This is my read READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E The IOBUFF is a full word pointer to the storage I obtain Here is my code LAR0,IODSECTLEN STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 LRR9,R6 USING IHADCB,R9 USING IODSECT,R6 * * LAR7,SYSADATA MVC 0(IODSECTLEN,R6),0(R7) * LAR7,SYSDCBE-SYSADATA(,R6) STR7,DCBDCBE LAR7,EX24LST-SYSADATA(,R6) STCM R7,B'0111',DCBEXLSA LAR7,ABND24-SYSADATA(,R6) STCM R7,B'0111',ABND24ADR-SYSADATA(R6) LAR7,ABENDRTN STR7,ABEND31-SYSADATA(,R6) * T L R0,=F'31996' * STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 *LAR7,IOAREA *STR7,IOPTR * STR7,IOBUFF MVC OPEN_LST(OPEN_LEN),OPEN_CON OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 * TMDCBOFLGS,DCBOFOPN BZEXITDEBG USING ASMADATA,R7 L R10,IOBUFF PROCADATA DS 0H READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E * CHECK ADATADECB * OPEN_CON OPEN (*-*,INPUT),MF=L,MODE=31 SYSADATA DCB DDNAME=SYSADATA,RECFM=VB,MACRF=RP,DSORG=PS,DCBE=SYSDCBE,EXX LST=EX24LST IHADECB DSECT=NO DECBLEN EQU *-DECB SYSDCBE DCBE EODAD=FIN,GETSIZE=YES,LOC=ANY,RMODE31=BUFF,SYNAD=SYNAD,VEX RSION=1 EX24LSTDS 0F ADATAEXLST DC AL1(EXLDCBAB+X'80') ABND24ADR DS AL3 ABND24 DS 0H L R15,ABEND31-SYSADATA(,R15) BSM 0,R15 ABEND31DS XL4 STROGE24 EQU *-EX24LST * IODSECT DSECT ADATADCB DSCL(DCBLNGBS) ADATADECB DSCL(DECBLEN) ADATADCBE DSCL(DCBELENV1) DS0F INST31DSCL(STROGE24) IODSECTLEN EQU *-IODSECT -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN -- For IBM-MAIN subscribe / signoff / archive access instructions, send email to lists...@listserv.ua.edu with the message: INFO IBM-MAIN
Re: BSAM READ AMODE 31 RMODE 31
Thanks MVC 0(IODSECTLEN,R6),SYSADATA LAR7,SYSDCBE STR7,DCBDCBE LAR7,EX24LST STCM R7,B'0111',DCBEXLSA LAR7,ABND24 STCM R7,B'0111',ABND24ADR-SYSADATA(R6) LAR7,ABENDRTN STR7,ABEND31-SYSADATA(,R6) -Original Message- From: IBM Mainframe Discussion List On Behalf Of Seymour J Metz Sent: Monday, March 20, 2023 11:07 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: BSAM READ AMODE 31 RMODE 31 That doesn't explain why you had LA R7 in places where you didn't need a pointer, LAR7,SYSADATA MVC 0(IODSECTLEN,R6),0(R7) Instead of the clearer MVC 0(IODSECTLEN,R6),SYSADATA Side issue are why you had separate moves for initialization instead of one big move and why you didn't just use symbolic names in your E-form macros instead of register form parameters with an otherwise extraneous LA. You might want to read up on dependent and labelled USING pseudo-ops. Only the WAIT service should be storing an RB address in an ECB; the ECB that you provide in the DECB should initially have a high bit of 0. Also, the completion code is 31 bits, and a lot of system services have codes with the lower 24 bits zero or containig a context for the code in bits 0-7. In particular, Many services use 7F in the high byte for normal completion. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Joseph Reichman [reichman...@gmail.com] Sent: Monday, March 20, 2023 10:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: BSAM READ AMODE 31 RMODE 31 Ok Thanks The reason I had the LA R7 I first had to get 24 bit storage for 1) dcb 2) decb though I realize it can live in 31 bit 3) decb it has to live in 31 bit 4) initial exit code has to be 24 bit Once moved I had to get the proper pointers 24 bit pointers for exlst , decb Let me corroborate I understand why I got the error After the read the first full word should be for the ECB so the low order should have an RB Or if it already posted should have the first byte x'40' followed by completion code which could be zeros what a saw in the ECB address was the first byte x'7F' Anyway I have to fix the read thank you > On Mar 20, 2023, at 9:52 AM, Binyamin Dissen wrote: > > First of all, the buffer address specified on READ MF=E is an RX > address, not an indirect address. > > So the read will overlay from IOBBUFF. That could cause strange > results, > > You would need > > L Rx,IOBUFF > READ ADATADECB,SF,(R6),(Rx),'S',MF=E > > > Also, I do not understand why you use IHADECB in your work area rather > than a > > READ label,SF,1-1,2-2,'S',MF=L > > which will reserve the storage of the appropriate size. > > I would recommend against altering the DECB directly, as the READ > macro takes care of it. > > Very strange to STORAGE OBTAIN,ADDR=(R7) followed by a LA R7 > > Back to the drawing board. > > On Mon, 20 Mar 2023 09:20:39 -0400 Joseph Reichman > > wrote: > > :> Hi > :> > :> > :> > :>I know Charles Mills answered my question nearly 4 weeks ago but I > would :>just like to corroborate the I/O areas for the read can be 31 > bit addressing :>only the dcb and decb need to be below the line :> > :>I am running AMODE 31 RMODE 31 :> :> :> :>The DCB is below the line > as well as the DECB :> :> :> :>My read returns ok :> :> :> :>After I > set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP :> > :> :> :>This the example that Charles sent to me does nt say anything > about the :>IOAREAS or pointers to the IOAREAS :> :> :> :>LA Rn,PDSDCB > :> :>ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB :> :>L Rn,address of > READ buffer :> :>READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB :> :> :> :> > :> :> :> :> :> :>This is my read :> :> :> :> READ > ADATADECB,SF,(R6),IOBUFF,'S',MF=E :> :> :> :>The IOBUFF is a full > word pointer to the storage I obtain :> :> :> :> :> :> Here is my > code :> :> :> > :> LAR0,IODSECTLEN > :> > :> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 > :> > :> LRR9,R6 > :> > :> USING IHADCB,R9 > :> > :> USING IODSECT
Re: BSAM READ AMODE 31 RMODE 31
You are correct sir Will make Binyamin changes and try it tonite > On Mar 20, 2023, at 11:07 AM, Seymour J Metz wrote: > > That doesn't explain why you had LA R7 in places where you didn't need a > pointer, > > LAR7,SYSADATA > MVC 0(IODSECTLEN,R6),0(R7) > > Instead of the clearer > > MVC 0(IODSECTLEN,R6),SYSADATA > > Side issue are why you had separate moves for initialization instead of one > big move and why you didn't just use symbolic names in your E-form macros > instead of register form parameters with an otherwise extraneous LA. > > You might want to read up on dependent and labelled USING pseudo-ops. > > Only the WAIT service should be storing an RB address in an ECB; the ECB that > you provide in the DECB should initially have a high bit of 0. Also, the > completion code is 31 bits, and a lot of system services have codes with the > lower 24 bits zero or containig a context for the code in bits 0-7. In > particular, Many services use 7F in the high byte for normal completion. > > > -- > Shmuel (Seymour J.) Metz > http://mason.gmu.edu/~smetz3 > > > From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of > Joseph Reichman [reichman...@gmail.com] > Sent: Monday, March 20, 2023 10:10 AM > To: IBM-MAIN@LISTSERV.UA.EDU > Subject: Re: BSAM READ AMODE 31 RMODE 31 > > Ok > > Thanks > The reason I had the LA R7 > > I first had to get 24 bit storage for 1) dcb 2) decb though I realize it can > live in 31 bit > 3) decb it has to live in 31 bit 4) initial exit code has to be 24 bit > > Once moved I had to get the proper pointers > 24 bit pointers for exlst , decb > > Let me corroborate I understand why I got the error > > After the read the first full word should be for the ECB so the low order > should have an RB > Or if it already posted should have the first byte x’40’ followed by > completion code which could be zeros what a saw in the ECB address was the > first byte x’7F’ > > Anyway I have to fix the read thank you > >> On Mar 20, 2023, at 9:52 AM, Binyamin Dissen >> wrote: >> >> First of all, the buffer address specified on READ MF=E is an RX address, not >> an indirect address. >> >> So the read will overlay from IOBBUFF. That could cause strange results, >> >> You would need >> >> L Rx,IOBUFF >> READ ADATADECB,SF,(R6),(Rx),'S',MF=E >> >> >> Also, I do not understand why you use IHADECB in your work area rather than a >> >> READ label,SF,1-1,2-2,'S',MF=L >> >> which will reserve the storage of the appropriate size. >> >> I would recommend against altering the DECB directly, as the READ macro takes >> care of it. >> >> Very strange to STORAGE OBTAIN,ADDR=(R7) followed by a LA R7 >> >> Back to the drawing board. >> >> On Mon, 20 Mar 2023 09:20:39 -0400 Joseph Reichman >> wrote: >> >> :> Hi >> :> >> :> >> :> >> :>I know Charles Mills answered my question nearly 4 weeks ago but I would >> :>just like to corroborate the I/O areas for the read can be 31 bit >> addressing >> :>only the dcb and decb need to be below the line >> :> >> :>I am running AMODE 31 RMODE 31 >> :> >> :> >> :> >> :>The DCB is below the line as well as the DECB >> :> >> :> >> :> >> :>My read returns ok >> :> >> :> >> :> >> :>After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP >> :> >> :> >> :> >> :>This the example that Charles sent to me does nt say anything about the >> :>IOAREAS or pointers to the IOAREAS >> :> >> :> >> :> >> :>LA Rn,PDSDCB >> :> >> :>ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB >> :> >> :>L Rn,address of READ buffer >> :> >> :>READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB >> :> >> :> >> :> >> :> >> :> >> :> >> :> >> :> >> :> >> :>This is my read >> :> >> :> >> :> >> :> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E >> :> >> :> >> :> >> :>The IOBUFF is a full word pointer to the storage I obtain >> :> >> :> >> :> >> :> >> :> >> :> Here is my code >> :> >> :> &g
Re: BSAM READ AMODE 31 RMODE 31
That doesn't explain why you had LA R7 in places where you didn't need a pointer, LAR7,SYSADATA MVC 0(IODSECTLEN,R6),0(R7) Instead of the clearer MVC 0(IODSECTLEN,R6),SYSADATA Side issue are why you had separate moves for initialization instead of one big move and why you didn't just use symbolic names in your E-form macros instead of register form parameters with an otherwise extraneous LA. You might want to read up on dependent and labelled USING pseudo-ops. Only the WAIT service should be storing an RB address in an ECB; the ECB that you provide in the DECB should initially have a high bit of 0. Also, the completion code is 31 bits, and a lot of system services have codes with the lower 24 bits zero or containig a context for the code in bits 0-7. In particular, Many services use 7F in the high byte for normal completion. -- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 From: IBM Mainframe Discussion List [IBM-MAIN@LISTSERV.UA.EDU] on behalf of Joseph Reichman [reichman...@gmail.com] Sent: Monday, March 20, 2023 10:10 AM To: IBM-MAIN@LISTSERV.UA.EDU Subject: Re: BSAM READ AMODE 31 RMODE 31 Ok Thanks The reason I had the LA R7 I first had to get 24 bit storage for 1) dcb 2) decb though I realize it can live in 31 bit 3) decb it has to live in 31 bit 4) initial exit code has to be 24 bit Once moved I had to get the proper pointers 24 bit pointers for exlst , decb Let me corroborate I understand why I got the error After the read the first full word should be for the ECB so the low order should have an RB Or if it already posted should have the first byte x’40’ followed by completion code which could be zeros what a saw in the ECB address was the first byte x’7F’ Anyway I have to fix the read thank you > On Mar 20, 2023, at 9:52 AM, Binyamin Dissen > wrote: > > First of all, the buffer address specified on READ MF=E is an RX address, not > an indirect address. > > So the read will overlay from IOBBUFF. That could cause strange results, > > You would need > > L Rx,IOBUFF > READ ADATADECB,SF,(R6),(Rx),'S',MF=E > > > Also, I do not understand why you use IHADECB in your work area rather than a > > READ label,SF,1-1,2-2,'S',MF=L > > which will reserve the storage of the appropriate size. > > I would recommend against altering the DECB directly, as the READ macro takes > care of it. > > Very strange to STORAGE OBTAIN,ADDR=(R7) followed by a LA R7 > > Back to the drawing board. > > On Mon, 20 Mar 2023 09:20:39 -0400 Joseph Reichman > wrote: > > :> Hi > :> > :> > :> > :>I know Charles Mills answered my question nearly 4 weeks ago but I would > :>just like to corroborate the I/O areas for the read can be 31 bit addressing > :>only the dcb and decb need to be below the line > :> > :>I am running AMODE 31 RMODE 31 > :> > :> > :> > :>The DCB is below the line as well as the DECB > :> > :> > :> > :>My read returns ok > :> > :> > :> > :>After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP > :> > :> > :> > :>This the example that Charles sent to me does nt say anything about the > :>IOAREAS or pointers to the IOAREAS > :> > :> > :> > :>LA Rn,PDSDCB > :> > :>ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB > :> > :>L Rn,address of READ buffer > :> > :>READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB > :> > :> > :> > :> > :> > :> > :> > :> > :> > :>This is my read > :> > :> > :> > :> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > :> > :> > :> > :>The IOBUFF is a full word pointer to the storage I obtain > :> > :> > :> > :> > :> > :> Here is my code > :> > :> > :> > :> LAR0,IODSECTLEN > :> > :> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 > :> > :> LRR9,R6 > :> > :> USING IHADCB,R9 > :> > :> USING IODSECT,R6 > :> > :> * > :> > :> * > :> > :> LAR7,SYSADATA > :> > :> MVC 0(IODSECTLEN,R6),0(R7) > :> > :> * > :> > :> LAR7,SYSDCBE-SYSADATA(,R6) > :> > :> STR7,DCBDCBE > :> > :> LAR7,EX24LST-SYSADATA(,R6) > :> > :> STCM R7,B'0111',DCBEXLSA > :> > :> LAR7,ABND24-SYSADATA(,R6) > :> > :> STCM R7,B'0111',ABND
Re: BSAM READ AMODE 31 RMODE 31
Ok Thanks The reason I had the LA R7 I first had to get 24 bit storage for 1) dcb 2) decb though I realize it can live in 31 bit 3) decb it has to live in 31 bit 4) initial exit code has to be 24 bit Once moved I had to get the proper pointers 24 bit pointers for exlst , decb Let me corroborate I understand why I got the error After the read the first full word should be for the ECB so the low order should have an RB Or if it already posted should have the first byte x’40’ followed by completion code which could be zeros what a saw in the ECB address was the first byte x’7F’ Anyway I have to fix the read thank you > On Mar 20, 2023, at 9:52 AM, Binyamin Dissen > wrote: > > First of all, the buffer address specified on READ MF=E is an RX address, not > an indirect address. > > So the read will overlay from IOBBUFF. That could cause strange results, > > You would need > > L Rx,IOBUFF > READ ADATADECB,SF,(R6),(Rx),'S',MF=E > > > Also, I do not understand why you use IHADECB in your work area rather than a > > READ label,SF,1-1,2-2,'S',MF=L > > which will reserve the storage of the appropriate size. > > I would recommend against altering the DECB directly, as the READ macro takes > care of it. > > Very strange to STORAGE OBTAIN,ADDR=(R7) followed by a LA R7 > > Back to the drawing board. > > On Mon, 20 Mar 2023 09:20:39 -0400 Joseph Reichman > wrote: > > :> Hi > :> > :> > :> > :>I know Charles Mills answered my question nearly 4 weeks ago but I would > :>just like to corroborate the I/O areas for the read can be 31 bit addressing > :>only the dcb and decb need to be below the line > :> > :>I am running AMODE 31 RMODE 31 > :> > :> > :> > :>The DCB is below the line as well as the DECB > :> > :> > :> > :>My read returns ok > :> > :> > :> > :>After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP > :> > :> > :> > :>This the example that Charles sent to me does nt say anything about the > :>IOAREAS or pointers to the IOAREAS > :> > :> > :> > :>LA Rn,PDSDCB > :> > :>ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB > :> > :>L Rn,address of READ buffer > :> > :>READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB > :> > :> > :> > :> > :> > :> > :> > :> > :> > :>This is my read > :> > :> > :> > :> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E > :> > :> > :> > :>The IOBUFF is a full word pointer to the storage I obtain > :> > :> > :> > :> > :> > :> Here is my code > :> > :> > :> > :> LAR0,IODSECTLEN > :> > :> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 > :> > :> LRR9,R6 > :> > :> USING IHADCB,R9 > :> > :> USING IODSECT,R6 > :> > :> * > :> > :> * > :> > :> LAR7,SYSADATA > :> > :> MVC 0(IODSECTLEN,R6),0(R7) > :> > :> * > :> > :> LAR7,SYSDCBE-SYSADATA(,R6) > :> > :> STR7,DCBDCBE > :> > :> LAR7,EX24LST-SYSADATA(,R6) > :> > :> STCM R7,B'0111',DCBEXLSA > :> > :> LAR7,ABND24-SYSADATA(,R6) > :> > :> STCM R7,B'0111',ABND24ADR-SYSADATA(R6) > :> > :> LAR7,ABENDRTN > :> > :> STR7,ABEND31-SYSADATA(,R6) > :> > :> * > :> > :>T L R0,=F'31996' > :> > :> * > :> > :> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 > :> > :> *LAR7,IOAREA > :> > :> *STR7,IOPTR > :> > :> * > :> > :> STR7,IOBUFF > :> > :> MVC OPEN_LST(OPEN_LEN),OPEN_CON > :> > :> OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 > :> > :> * > :> > :> TMDCBOFLGS,DCBOFOPN > :> > :> BZEXITDEBG
Re: BSAM READ AMODE 31 RMODE 31
First of all, the buffer address specified on READ MF=E is an RX address, not an indirect address. So the read will overlay from IOBBUFF. That could cause strange results, You would need L Rx,IOBUFF READ ADATADECB,SF,(R6),(Rx),'S',MF=E Also, I do not understand why you use IHADECB in your work area rather than a READ label,SF,1-1,2-2,'S',MF=L which will reserve the storage of the appropriate size. I would recommend against altering the DECB directly, as the READ macro takes care of it. Very strange to STORAGE OBTAIN,ADDR=(R7) followed by a LA R7 Back to the drawing board. On Mon, 20 Mar 2023 09:20:39 -0400 Joseph Reichman wrote: :> Hi :> :> :> :>I know Charles Mills answered my question nearly 4 weeks ago but I would :>just like to corroborate the I/O areas for the read can be 31 bit addressing :>only the dcb and decb need to be below the line :> :>I am running AMODE 31 RMODE 31 :> :> :> :>The DCB is below the line as well as the DECB :> :> :> :>My read returns ok :> :> :> :>After I set a beak point after the CHECK I get As S0D9 ABEND with a SDUMP :> :> :> :>This the example that Charles sent to me does nt say anything about the :>IOAREAS or pointers to the IOAREAS :> :> :> :>LA Rn,PDSDCB :> :>ST Rn,PDSDECB+8 PLACE DCB ADDRESS IN DECB :> :>L Rn,address of READ buffer :> :>READ PDSDECB,SF,,(Rn),MF=E CHECK PDSDECB :> :> :> :> :> :> :> :> :> :>This is my read :> :> :> :> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E :> :> :> :>The IOBUFF is a full word pointer to the storage I obtain :> :> :> :> :> :> Here is my code :> :> :> :> LAR0,IODSECTLEN :> :> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R6),LOC=BELOW,SP=0 :> :> LRR9,R6 :> :> USING IHADCB,R9 :> :> USING IODSECT,R6 :> :> * :> :> * :> :> LAR7,SYSADATA :> :> MVC 0(IODSECTLEN,R6),0(R7) :> :> * :> :> LAR7,SYSDCBE-SYSADATA(,R6) :> :> STR7,DCBDCBE :> :> LAR7,EX24LST-SYSADATA(,R6) :> :> STCM R7,B'0111',DCBEXLSA :> :> LAR7,ABND24-SYSADATA(,R6) :> :> STCM R7,B'0111',ABND24ADR-SYSADATA(R6) :> :> LAR7,ABENDRTN :> :> STR7,ABEND31-SYSADATA(,R6) :> :> * :> :>T L R0,=F'31996' :> :> * :> :> STORAGE OBTAIN,LENGTH=(R0),ADDR=(R7),LOC=RES,SP=0 :> :> *LAR7,IOAREA :> :> *STR7,IOPTR :> :> * :> :> STR7,IOBUFF :> :> MVC OPEN_LST(OPEN_LEN),OPEN_CON :> :> OPEN ((R6),INPUT),MF=(E,OPEN_LST),MODE=31 :> :> * :> :> TMDCBOFLGS,DCBOFOPN :> :> BZEXITDEBG :> :> USING ASMADATA,R7 :> :> L R10,IOBUFF :> :> PROCADATA DS 0H :> :> READ ADATADECB,SF,(R6),IOBUFF,'S',MF=E :> :> * :> :> CHECK ADATADECB :> :> * :> :>OPEN_CON OPEN (*-*,INPUT),MF=L,MODE=31 :> :>SYSADATA DCB DDNAME=SYSADATA,RECFM=VB,MACRF=RP,DSORG=PS,DCBE=SYSDCBE,EXX :> :> LST=EX24LST :> :> :> :> IHADECB DSECT=NO :> :>DECBLEN EQU *-DECB :> :>SYSDCBE DCBE EO
