Google
 

Trailing-Edge - PDP-10 Archives - bb-bt99h-bb - ihsv01.d08
There are 2 other files named ihsv01.d08 in the archive. Click here to see a list.
                 EDIT DESCRIPTIONS FOR IBMCOM-10-HASP-V1                        
  
  
                             EDIT 006    FOR D60SPD
  
  
  
  
[SYMPTOM]
  
Garbled filespecs if the default directory has SFDs.
  
  
[DIAGNOSIS]
  
Routine GETPTH increments T1 once for each SFD level  present  in
the  default  directory.   However,  routine S$FILE expects T1 to
point to DEVDEV upon entry, which it does not after having called
GETPTH.
  
  
[CURE]
  
Point T1 at DEVDEV before calling S$FILE.
********************************************************************************
  
  
                             EDIT 007    FOR D60SPD
  
[SYMPTOM]
  
Errors during the compilation of D60SPD with TOPS-10 Version 7.02
MACTEN.UNV
  
  
  
[DIAGNOSIS]
  
Redefination of MOVX macro is incompatible with field image D60SPD
  
  
  
[C]	Temporary
  
Search the version of MACTEN that is compatible with D60SPD
until MOVX can be made to work properly within D60SPD.
  
  
  
********************************************************************************
  
  
                             EDIT 034    FOR DN60
  
[SYMPTOM]
Spurious crlf sequences appear in output returning from
an IBM host.
  
[DIAGNOSIS]
Additional end of line processing with edit 33 installed
may introduce an extra crlf in some situations.
  
  
[CURE]
Remove edit 33 from module XL3780.P11 of the Tops-10 version 4
2780/3780 ET product.
********************************************************************************
  
  
                             EDIT 351    FOR IBMSPL
  
This Edit has no associated Symptom, Diagnosis, and Cure file.
********************************************************************************
  
  
                             EDIT 352    FOR IBMSPL
  
This Edit has no associated Symptom, Diagnosis, and Cure file.
********************************************************************************
  
  
                             EDIT 353    FOR IBMSPL
  
  
  
  
[SYMPTOM]
  
IBMSPL gets accounting error  5  when  executing  in  DEBUG  mode
(FTDEBUG non-zero).
  
  
[DIAGNOSIS]
  
IBMSPL executes a DAEMON UUO to make entries  in  the  FACT  file
(function  code  .FACT).   If IBMSPL is being debugged from other
than [1,2] then the DAEMON UUO will take the error return.
  
  
[CURE]
  
Conditionalize routine DOFACT to not execute any accounting  code
if FTDEBUG is non-zero.
********************************************************************************
  
  
                             EDIT 354    FOR IBMSPL
  
  
  
  
[SYMPTOM]
  
IBMSPL gets ILMs when FTFACT is non-zero.
  
  
[DIAGNOSIS]
  
In routine DOFACT the LOAD macro is used incorrectly to load  the
queue  name,  the  sequence  number  and  the  number  of records
processed.  Since the DATAST, $ and $.   macros  produce  symbols
which  contain  an  implicit  index register, references to these
symbols with the LOAD and STORE macros must leave their  position
fields blank.
  
  
[CURE]
  
Change the code in DOFACT to  use  the  LOAD  macro  to  properly
reference T.QNM,T.SEQ and T.NRS.
********************************************************************************
  
  
                             EDIT 355    FOR IBMSPL
  
  
  
  
[SYMPTOM]
  
?Illegal address in UUO at user PC xxxxxx
  
  
[DIAGNOSIS]
  
The UUO which is causing  the  problem  is  the  QUEUE.   UUO  at
WRTUSG:.   IBMSPL uses the USTXT.  macro to create anentry in the
argument block for the QUEUE.  UUO incorrectly.  There are  three
occurances of:
  
        USTXT.  (<-1,,[ASCIZ / /]>)     ;SYSTEM TEXT
  
in the code.  The macro expands to two words which look something
like:
  
        XWD     10,112
        Z       @string_address(17)
  
The indirect bit and the index  AC  (17,  in  this  case)  appear
because  the  argument  is masked with 37777777 when the macro is
expanded.  This is clearly an incorrect address.  An  interesting
aside  is  that  under certain circumstances this address, though
incorrect, may indeed be legal in the sense that it is within the
confines of IBMSPL's virtual address space.
  
  
[CURE]
  
Correct the usage of the USTXT.  macro in IBMSPL.  This will  fix
IBMSPL such that it no longer generates incorrect addresses.
********************************************************************************
  
  
                             EDIT 356    FOR IBMSPL
  
  
  
  
[SYMPTOM]
  
Users are able to submit any  file  on  the  machine  to  an  IBM
system.
  
  
[DIAGNOSIS]
  
IBMSPL  uses  the  short  form  FOB.   Thus,   F%IOPN   does   no
"in-your-behalf" access checking.
  
  
[CURE]
  
Teach IBMSPL about the long form FOB so that F%IOPN can do access
checking.
********************************************************************************
  
  
                             EDIT 357    FOR IBMSPL
  
  
  
  
[SYMPTOM]
  
Wrong version printed at compile time.
  
  
[DIAGNOSIS]
  
Off-by-one in the VOUTX macro.
  
  
[CURE]
  
Replace the "A" with "@".
********************************************************************************
  
  
                             EDIT 360    FOR IBMSPL
  
[SYMPTOM]
  
Non observed
  
  
[DIAGNOSIS]
  
Test for Line up should occur before test for error on line.
  
  
[CURE]
  
Reverse the Line Up and Line Gone Away tests in the signoff process.
  
********************************************************************************
  
  
                             EDIT 361    FOR IBMSPL
  
[SYMPTOM]
  
Ibmspl gets accounting error when in debug mode.
  
  
[DIAGNOSIS]
  
Edit 353 fixed one case to prevent accounting in debug mode (FTDEBUG on)
Ibmspl can be debugged with FTDEBUG assembly time switch, or with
DEBUGW set to non-zero at run time.
  
  
[CURE]
  
Extend edit 353 to prevent accounting being done if DEBUGW is on.
  
********************************************************************************
  
  
  
END OF  IBMCOM-10-HASP-V1