View previous topic :: View next topic |
Author |
Message |
stfark1
Joined: 02 Sep 2008 Posts: 222
|
Posted: Wed Jun 22, 2022 5:33 pm Post subject: Strange Entry |
|
|
When running the debugger(64) on a program with ENTRY points a activating "step in" in the debugger, the system does not "step in", just "step over", thus cannot tell what is happening inside of the entry point. See code below, Sid Kraft
C MAIN ROUTINE TO DETERMINE ENTRY POINTS
COMMON/ENTVARS/IENTR1,IENTR2,IENTR3,IENTR4,ISBSET
IENTR1=0
IENTR2=0
IENTR3=0
IENTR4=0
CALL SUBNT1
CALL SUBNT2
CALL SUBNT3
CALL SUBNT4
C ENTER THE CALL NUMBER
C WRITE(6,100)
C100 FORMAT(1X,"ENTER THE SUBSET TO CALL AS FOLLOWS:")
C WRITE(6,110)
C110 FORMAT(1X,"1=MAIN,2=ENTRY1,3=ENTRY2,3=ENTRY3,4=ENTRY4")
C READ(5,120) ISBSET
C120 FORMAT(I4)
C NOTE INPUT AS FOLLOWS:
C 1 = MAIN PROGRAM
C 2 = ENTRY 1
C 3 = ENTRY 2
C 4 = ENTRY 3
C 5 = ENTRY 4
CALL SUBENT
END
C Entry Routine, called from Main MNent
SUBROUTINE SUBENT
COMMON/ENTVARS/ IENTR1,IENTR2,IENTR3,IENTR4,ISBSET
C GOTO(50,150,250,350,450),ISBSET
C50 WRITE(6,100)
C100 FORMAT(1x,"Got To Main Entry")
C RETURN
ENTRY SUBNT1
150 WRITE(6,200)
200 FORMAT(1x,"Got To Subnt1")
RETURN
ENTRY SUBNT2
250 WRITE(6,300)
300 FORMAT(1X,"GOT TO SUBNT2")
RETURN
ENTRY SUBNT3
350 WRITE(6,400)
400 FORMAT(1X,"GOT TO SUBNT3")
RETURN
ENTRY SUBNT4
450 WRITE(6,500)
500 FORMAT(1X,"GOTO TO SUBNT4")
RETURN
END
Please advise |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 8017 Location: Salford, UK
|
Posted: Wed Jun 22, 2022 6:11 pm Post subject: |
|
|
stfark1
Are you writing new code? If you are then you should not be using ENTRY points because the Standard classifies them as "depreciated" i.e. obsolete.
If you are using a vast amount of old code that uses ENTRY points then, if the compiler or debugger was failing in some way then it would be reasonable to ask for a fix. |
|
Back to top |
|
|
stfark1
Joined: 02 Sep 2008 Posts: 222
|
Posted: Wed Jun 22, 2022 8:04 pm Post subject: Strange Entry |
|
|
Paul: As I mentioned in previous entries, have a very large program consisting of 30 Fortran Routines, several thousand lines of code in each routine with 41 entry points. to convert all entry points to "branching" code with that routine that contains the entry points will be a very large task. This is why I am asking for a fix rather than a total conversion. Do not understand why the 64 bit system was created to not include standard Fortran statements, i.e. Entry points is had to imagine. Anyway, if there is a work around, please let me know, Sid Kraft |
|
Back to top |
|
|
stfark1
Joined: 02 Sep 2008 Posts: 222
|
Posted: Thu Jun 23, 2022 12:35 am Post subject: Strange Entry |
|
|
Paul: As a matter of curiosity, re-compiled all using debug 32 and the ENTRY XXXX works fine, when step into or step over, both work as they were designed to work. I did get the large program to work under debug 32 but really wanted to use the debug 64 system as it has later modifications. Sid Kraft |
|
Back to top |
|
|
mecej4
Joined: 31 Oct 2006 Posts: 1896
|
Posted: Thu Jun 23, 2022 2:51 am Post subject: |
|
|
Do you have subroutines/functions with ENTRYs such that some of those ENTRY points have argument lists different from the argument list of the containing SUBROUTINE/FUNCTION declaration? |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 8017 Location: Salford, UK
|
Posted: Thu Jun 23, 2022 6:38 am Post subject: |
|
|
Sid
FTN95 does provide ENTRY points for 64 bit code and your experience suggests that there may be a related fault in either FTN95 or SDBG64.
I will add this to the list of issues that need investigating. I assume that the code you have provided above demonstrates a fault. |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 8017 Location: Salford, UK
|
Posted: Thu Jun 23, 2022 6:56 am Post subject: |
|
|
Sid
I do not get the same experience when stepping through the code provided above. I am using FTN95 v8.90 which is currently only available to supported users.
If you are using the personal edition of FTN95 then could try updating to v8.84 that can be downloaded via the link below.
I do get a minor fault when entering SUBNT1 in that the WRITE statement is not provided with a break point. The other routines do not exhibit this fault.
http://forums.silverfrost.com/viewtopic.php?t=4245 |
|
Back to top |
|
|
stfark1
Joined: 02 Sep 2008 Posts: 222
|
Posted: Thu Jun 23, 2022 4:11 pm Post subject: Strange Entry |
|
|
Paul: Per your request, I will download the latest version available, V8.84 to restore to the latest version. I have an entry in my Program File(x86) called Silverfrost, I assume that I should delete this folder and then download the update in order to completely restore all, please verify. Sid Kraft |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 8017 Location: Salford, UK
|
Posted: Thu Jun 23, 2022 4:39 pm Post subject: |
|
|
Sid
The downloads contain components not the full release so you should not delete the Silverfrost folder.
If you are not sure how to use the downloads then it would be better to wait for the next full release of the personal edition. |
|
Back to top |
|
|
stfark1
Joined: 02 Sep 2008 Posts: 222
|
Posted: Thu Jun 23, 2022 5:00 pm Post subject: Strange Entry |
|
|
Paul: Not sure how to use, is it difficult, i.e. just download and extract the debugger update, when I extract will this automatically update the elements in the Silverfrost folder? Hate to wait for the full release, not sure when this will be and how do I know? Sid Kraft |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 8017 Location: Salford, UK
|
Posted: Fri Jun 24, 2022 6:38 am Post subject: |
|
|
Sid
Maybe it would help if you could find someone to do a Zoom session with you whilst you go through the downloading. Unfortunately I don't have the time to help users in this way. |
|
Back to top |
|
|
|