Login
Username:

Password:

Remember me



Lost Password?

Register now!

Sections

Who's Online
26 user(s) are online (13 user(s) are browsing Forums)

Members: 0
Guests: 26

more...

Headlines

 
  Register To Post
« 1 2 3 4 (5)

Re: SpotLess tests need it to find out one bug
Just can't stay away
Just can't stay away


See User information
@rjd324

Since you are using SimpleDebug, I am guessing, that you are using the version of Spotless available at os4depot. This version is obsolete and should probably be removed (please remind me). There are some links to newer executables in this thread, that all lead to the same file location at google drive. These are beta versions but are definitely better for testing at this point than the old version.

Anyways, the problem you mention with having a crash on the X5000 and not on the X1000 is very strange. I am guessing, that it has to do with the standard library version or some other component in that category, since your stacktrace shows a crash at just a std::vector constructor (so, there is a bug, but it only generates a crash on one platform due to other software on the machine).

Go to top
Re: SpotLess tests need it to find out one bug
Not too shy to talk
Not too shy to talk


See User information
Yes. I am using the one from os4Depot.

Now, I replaced my "spotless" binary with this one: https://drive.google.com/file/d/1SD-sX ... At7fyC3O/view?usp=sharing (this is the latest link I could find). File size: 8699013.

I then followed @kas1e's steps from the first post in this thread.

One of those steps is to load his "test" executable. When I do this on both the X1 and X5 machines I see the assembly instructions. From there, I place the BP on line 30. When I hit start on the X5 though I see this in the "Disassembler" tab/section:

<not a readable address>


And, in the "Console" tab/section:

Source file selectedtest.c
At 
break : at line 0


I can keep on clicking "Start" but nothing happens.

The difference on the X1 machine is that when I click start the PC does get to line 30 as expected, and I can still see the instructions.

===

I am not sure if there is an equivalent command to ldd on the Amiga? I would like to know which shared objects have been linked with so that I can compare the versions on the X1 and X5 machine. All I know is that I have both the latest SDK versions on both machines and that I use https://github.com/sodero/adtools/rele ... cc11.1.0-20210531-743.lha those tools and replace that whole thing as my "gcc" directory since I use that instead of the 4.2.4 version that comes with the SDK today. So, if it is a linking with some libc/++ shared object then both the versions on the X1 X5 should be the same.

If liberty means anything at all, it means the right to tell people what they do not want to hear.
George Orwell.
Go to top
Re: SpotLess tests need it to find out one bug
Home away from home
Home away from home


See User information
@rjd324

objdump

(NutsAboutAmiga)

Basilisk II for AmigaOS4
AmigaInputAnywhere
Excalibur
and other tools and apps.
Go to top
Re: SpotLess tests need it to find out one bug
Just can't stay away
Just can't stay away


See User information
@rjd324

Are you running pulic update 2? If so, then that is the reason.

Go to top
Re: SpotLess tests need it to find out one bug
Not too shy to talk
Not too shy to talk


See User information
Hi,

public update 2? I do not know what you are referring to. Surely not FE update 2.

If liberty means anything at all, it means the right to tell people what they do not want to hear.
George Orwell.
Go to top
Re: SpotLess tests need it to find out one bug
Just can't stay away
Just can't stay away


See User information
@rjd324

Yes. That's exactly what I am referring to.

Go to top
Re: SpotLess tests need it to find out one bug
Not too shy to talk
Not too shy to talk


See User information
Yes, I run Amiga OS4.1 Final Edition Update 2 + the two hotfixes, on both X1 and X5. I do not have a crash on the X1, but I have a crash on the X5.

If liberty means anything at all, it means the right to tell people what they do not want to hear.
George Orwell.
Go to top
Re: SpotLess tests need it to find out one bug
Just can't stay away
Just can't stay away


See User information
@rjd324

Ok, cool. That was all I needed to know.

Apparently the software (Spotless) works on some setups pretty much perfectly (your X1000 setup and my X5000 setup). Now, there are a lot of different versions of system components out there, so the short version of the story is, that we have working performance on some combination of system components and faulty performance on others.

The long version would be a lengthy explanation of why the different combinations work or not. This story is of course hard to write, especially when I as the only active developer only have access to an already functioning setup. In your case I can only comment, that Spotless was known to not function properly on public update 2. Whether or not the functioning on your X5000 setup would be fixed by updating to available beta software in the kernel and other places would only be found out by actually updating.

Go to top

  Register To Post
« 1 2 3 4 (5)

 




Currently Active Users Viewing This Thread: 1 ( 0 members and 1 Anonymous Users )




Powered by XOOPS 2.0 © 2001-2016 The XOOPS Project