forums.silverfrost.com Forum Index forums.silverfrost.com
Welcome to the Silverfrost forums
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Compiler offset addresses missing(?)

 
Post new topic   Reply to topic    forums.silverfrost.com Forum Index -> 64-bit
View previous topic :: View next topic  
Author Message
wahorger



Joined: 13 Oct 2014
Posts: 1214
Location: Morrison, CO, USA

PostPosted: Sat Jan 16, 2021 5:48 pm    Post subject: Compiler offset addresses missing(?) Reply with quote

I am doing some debug on the 64-bit code, looking for issues of my own making. I encountered an error that give me an error:
Quote:
Argument number 2 of WINIO@ (continuation 2)
should be a 32 bit integer at address 1c0088e9

Within file C-MASTER.exe
in NEWMAIN at address 234
in CMAIN at address d06


My issue: I have the Listing file, and was looking for the offset, but found the following:

Quote:
0138 subroutine newmain(init_in_sc) AT f66
0139 include 'my_windows.ins' AT f66
0140 C_EXTERNAL CLEARWIN_PARAM@ '__clearwin_param'(INSTRING,INSTRING) AT f66
0141 include 'cmaster_routines.ins' AT f66
.
.
.
To the executable code
.
.
.
0190 screen_title = 'C-Master' AT f66
0191 AT f66
0192 AT f66
0193 call CLEARTYPE_WINDOW_PARAMETERS(local_window_parameters) AT f66
0194 if(run_once) then AT f66
0195 run_once = .false. AT f66
0196 call set_all_max_lines@(6000L) AT f66
0197 call flush_memory_buffers ! send lingering internal data to the appropriate file(s) AT f66
0198 CALL CLEARTYPE_WINDOW_PARAMETERS(local_window_parameters) AT f66
0199 local_window_parameters%sc_function_address = loc(main_window_one_time_init) ! specialized init for the first window in the "set" AT f66
0200 endif AT f66


The offset for each statement does not change. I looked at a number of other compilations in 64-bit, and the same condition exists, where the offset of the line of executable code is incorrect. I do like the indication of the continuation line being in error, but I cannot tell if it was a lower level routine that is being also counted, or if it is in the code as indicated. If the latter, this makes no sense, so I am confused.
Back to top
View user's profile Send private message Visit poster's website
PaulLaidler
Site Admin


Joined: 21 Feb 2005
Posts: 7912
Location: Salford, UK

PostPosted: Sat Jan 16, 2021 8:37 pm    Post subject: Reply with quote

This does suggest that /LIST is not working properly. I will put this down for investigation.

In the meantime try /EXPLIST to see if you can work out the offsets from the assembler listing.
Back to top
View user's profile Send private message AIM Address
wahorger



Joined: 13 Oct 2014
Posts: 1214
Location: Morrison, CO, USA

PostPosted: Sun Jan 17, 2021 1:22 am    Post subject: Reply with quote

It's a mixed bag.

Yes, /EXPlist does give me a better idea of where it is.

Unfortunately, the assembly code is physically separated from the actual source code. So while I can use this, it is a difficult slog to find the appropriate line in a set of lines all of whom are created similarly and differ in only one or two elements.

Bill
Back to top
View user's profile Send private message Visit poster's website
PaulLaidler
Site Admin


Joined: 21 Feb 2005
Posts: 7912
Location: Salford, UK

PostPosted: Sun Jan 17, 2021 9:33 am    Post subject: Reply with quote

OK. If you use /DEBUG then that may provide a line number for the failure.
Back to top
View user's profile Send private message AIM Address
wahorger



Joined: 13 Oct 2014
Posts: 1214
Location: Morrison, CO, USA

PostPosted: Sun Jan 17, 2021 5:05 pm    Post subject: Reply with quote

Yes, /DEBUG appears to give the line number within the file.

Just an FYI...

It might just be my system, but if I have /EXPLIST turned on as well, compilation is glacially slow.

Going back to /LIST goes faster (most of the time), but still really slow.

I modified the MAKE to allow me to select whether or not a listing file is created. For the production version, I'll have to balance speed with details.
Back to top
View user's profile Send private message Visit poster's website
PaulLaidler
Site Admin


Joined: 21 Feb 2005
Posts: 7912
Location: Salford, UK

PostPosted: Mon Jan 18, 2021 12:40 pm    Post subject: Reply with quote

The behaviour of FTN95 /64 /LIST has been greatly improved (maybe even fixed) for the next release. The same fix also provides the correct alignment for /EXPLIST.

In the meantime the offsets are only provided when /DEBUG is also applied.
Back to top
View user's profile Send private message AIM Address
Display posts from previous:   
Post new topic   Reply to topic    forums.silverfrost.com Forum Index -> 64-bit All times are GMT + 1 Hour
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group