Re: [hatari-devel] Bugs with memory snapshot saving from SDL GUI

[Eero Tamminen <oak@xxxxxxxxxxxxxx>]

Hi,

On 15.9.2022 1.55, Chris Jenkins wrote:
> > * Dialog not being shown when it should (or if SDL supports multiple
> > layers, not on top layer)
>
> I just replied to the thread in atari-forum about this last one. Pasting
> the same text here for visibility in this thread:
>
> I think what's happening here is that Hatari only actually attempts to
> write the snapshot to disk when emulation is unpaused, which in the case of
> hte SDL GUI happens when the GUI is closed. The relevant code is in
> MemorySnapShot_OpenFile in Hatari's source tree, which ultimately gets
> called from the UAE code, though I confess that I don't understand it all
> yet.
>
> I'm not clear whether this is the cause of the issues reported in this
> thread though.

Ahaa!

Debug UI calls MemorySnapShot_Capture_Immediate(), other cases do not:
--------------------------------------------
$ git grep '[^_]MemorySnapShot_Capture[(_]' | grep -F .c:
src/cpu/hatari-glue.c:	MemorySnapShot_Capture_Do ();
src/gui-osx/SDLMain.m:	 
MemorySnapShot_Capture(ConfigureParams.Memory.szMemoryCaptureFileName, 
TRUE);
src/debug/debugui.c:		MemorySnapShot_Capture_Immediate(file, true);
src/gui-sdl/dlgMemory.c:			 
MemorySnapShot_Capture(ConfigureParams.Memory.szMemoryCaptureFileName, 
true);
src/main.c:	 
MemorySnapShot_Capture(ConfigureParams.Memory.szAutoSaveFileName, false);
src/memorySnapShot.c:void MemorySnapShot_Capture(const char 
*pszFileName, bool bConfirm)
src/memorySnapShot.c:	/* Make a temporary copy of the parameters for 
MemorySnapShot_Capture_Do() */
src/memorySnapShot.c:void MemorySnapShot_Capture_Immediate(const char 
*pszFileName, bool bConfirm)
src/memorySnapShot.c:	/* Make a temporary copy of the parameters for 
MemorySnapShot_Capture_Do() */
src/memorySnapShot.c:	MemorySnapShot_Capture_Do ();
src/memorySnapShot.c:void MemorySnapShot_Capture_Do(void)
src/shortcut.c:	 
MemorySnapShot_Capture(ConfigureParams.Memory.szMemoryCaptureFileName, 
true);
--------------------------------------------

IMHO this does not need so many global functions, one with 
immediate/delayed bool would be enough, instead of having two, which 
would always checks file overwrite before setting up memory capture.

(If user wants to select another file instead overwriting existing one, 
he cannot do that with the delayed overwrite question.)


After that is fixed, there's still the alert level issue.  Question 
dialogs should be shown *always*.


> I'm not easily able to try to reproduce this issue myself because I'm using
> a Mac but I might try getting Hatari working on my Raspberry Pi if I get
> the chance and see if any of this stuff happens there.

According to sources, Mac GUI is also using the non-immediate memory 
capture function, so it should also be suffering of the too-late 
overwrite question.


	- Eero