View previous topic :: View next topic |
Author |
Message |
DanRRight
Joined: 10 Mar 2008 Posts: 2826 Location: South Pole, Antarctica
|
Posted: Thu Jan 02, 2020 9:25 pm Post subject: Please use /check with /debug or /full_debug |
|
|
Folks, I encourage all of you to compile all your tasks with such combinations of compilation switches
/check /debug
/check /full_debug
/check /debug /undef
I still sometimes have crash either during compilation at run time or debugging with sdbg64 if I use /check.
If we all find small demo examples which crash then Silverfrost might faster find the reason.
Sending to Silverfrost my large code with instructions what's to do with it would be a PITA for me and Silverfrost. Also I'm still guessing if such crashes in my cases are due to very large requirements on memory 60GB or more or something because only ones somebody reported this (probably that was mecej4) |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 7931 Location: Salford, UK
|
Posted: Fri Jan 03, 2020 9:13 am Post subject: |
|
|
/check includes /debug.
/undef includes /check.
/checkmate includes /undef.
So most users can use /checkmate but only for development and testing. There are high overheads. |
|
Back to top |
|
|
John-Silver
Joined: 30 Jul 2013 Posts: 1520 Location: Aerospace Valley
|
Posted: Fri Jan 03, 2020 12:58 pm Post subject: |
|
|
... or as it says in the definition of /CHECKMATE:
Quote: | /CHECKMATE
A synonym for /FULL_UNDEF. /FULL_UNDEF implies /UNDEF which in turn implies /CHECK. |
The documentation for checkmate here:
https://www.silverfrost.com/15/ftn95/checkmate_the_world's_best_runtime_checking.aspx
... imo probably neds a refresh (update).
From what I see there are 2 lists of things listed that CHECKMATE does, the lower list of 11 pointés seems to duplicate (in shortened form) some of the things explained in more detail in the upper list of 8 points.
For me it's a little confusing.
Nowhere is it stated that on uses '/CHECKMATE' to activate it !
Some examples of the possible combinations as mentioned in earlier comments here, should definately be documented somewhere and cross-referenced here in the CHECKMATE documentation in some way.
The correspondance/useage with /DEBUG should also be clarified somewhere too.
At the moment it's a bit of a 'black art' which is probably why people avoid it. Which is a bit of a shame.
Maybe it's a topic crying out for another good Video Guide Paul ?
Paul, as for your:
Quote: | So most users can use /checkmate ... |
which users can't ? , and why? _________________ ''Computers (HAL and MARVIN excepted) are incredibly rigid. They question nothing. Especially input data.Human beings are incredibly trusting of computers and don't check input data. Together cocking up even the simplest calculation ... " |
|
Back to top |
|
|
John-Silver
Joined: 30 Jul 2013 Posts: 1520 Location: Aerospace Valley
|
Posted: Fri Jan 03, 2020 1:08 pm Post subject: |
|
|
Dan,
apart from not understanding why you and Silverfrost would be incited to eat (I assume lot's of) kebabs a a result ,
your:
Quote: | If we all find small demo examples which crash then Silverfrost might faster find the reason. |
... made me laugh as it can be easily read as a call to arms to bring Silverfrost to it's knees !!! LOL
Which I realise of course it isn't, but more a call to Silverfrost themselves to more proactively advertise and encourage it's use. _________________ ''Computers (HAL and MARVIN excepted) are incredibly rigid. They question nothing. Especially input data.Human beings are incredibly trusting of computers and don't check input data. Together cocking up even the simplest calculation ... " |
|
Back to top |
|
|
Kenneth_Smith
Joined: 18 May 2012 Posts: 697 Location: Hamilton, Lanarkshire, Scotland.
|
|
Back to top |
|
|
DanRRight
Joined: 10 Mar 2008 Posts: 2826 Location: South Pole, Antarctica
|
Posted: Fri Jan 03, 2020 10:54 pm Post subject: |
|
|
Paul,
If /check includes /debug then is this an error when applying these two compilation switches
/check /debug
or it is just a redundant causing no harm?
John,
In this post we tried kind of "multiusers multitasking": the all users orchestrated massive bug hunt
One thing I found is that users tend not to report the bugs, they always get workarounds instead and wait others will report. So how we can make any software to be as clean as possible for as short time as possible ? Will see if activating all may help |
|
Back to top |
|
|
PaulLaidler Site Admin
Joined: 21 Feb 2005 Posts: 7931 Location: Salford, UK
|
Posted: Sat Jan 04, 2020 10:36 am Post subject: |
|
|
Dan
It is redundant causing no harm. |
|
Back to top |
|
|
|