@trixie I just found some time to read the latest of your blog posts. Thank you for sharing all this information. What I would like to ask is if you save in Undo only the part that is changed or the whole sample in its previous state. I guess this would make a huge difference in big samples and the used amount of memory, right?
@Trixie Nice read about ExMem. Wish it can be used a little bit more transparent for developers (like, you just replace malloc() with it, and be done :) )
@trixie Finally got a chance to read the blog. Thanks for writing the blogs. They are enjoyable and informative. Now to get some time to try out the undo feature. Rave will now be my "go to" sound editor.
@Dan Not sure if that correct topic to make a bugreports (and if it bug in the Rave at all, and not on my setup), but: recently i got x1000 and trying a rave on it (FE + update1/update2 + gadgets/classes/libs from Enhancer2.2 on top).
Now, almost all the time (but not all the time), when i run Rave, then choose Settings in RMB menu, and then hit "File Requester", then i crashes with such crashlog:
54.53 (17.7.2022) AmigaOne X1000 release
Machine model: 6 (AmigaOne X1000)
Dump of context at 0xDF8B7000
Trap type: DSI exception
Current kernel stack pointer: 0x0293FF00
DSISR: 40000000 DAR: B49CF200
No matching page found
Machine State (raw): 0x100000000200B030
Machine State (verbose): [Hyper] [ExtInt on] [Super] [IAT on] [DAT on]
Instruction pointer: in module kernel+0x00053FEC (0x02053FEC)
Crashed process: Rave (0x61BB8C00)
DSI verbose error description: Access to address 0xB49CF200 not found in hash or BAT (page fault)
Access was a load operation
0: 02054244 52E15400 00000000 61F11800 52E1660C 00000010 0000000E 0000000B
8: 61BB8BF0 61BB8BF4 52E1660C 52E16220 48224824 5F6A3E3C 52E15930 5C74D334
16: 5E970260 0000020A 61F6955E 00000000 00000000 00000000 00000004 52E161A0
24: 00000000 62297730 00000000 5C74CC5C 00000006 00000008 00000008 52E165FC
CR: 28224828 XER: 20000000 CTR: 02032478 LR: 02054244
@Ktadd Intersting .. You on public update2, right ? I am also on public update2, just "only" latest beta kernel + latest enhancer classes.
I retested again , and basically just going to settings, and go through "Display/File requester/Advanced" cause a crash. I.e. you can run settings, adn the click on "Advanced" to have the same crash.
@Dan Can you point out plz which exactly gadgets/classes Rave uses, so i can check maybe i have some beta enhancer gadgets which cause this..
Yes, I'm running latest public OS4 and Enhancer stuff with all updates from "Updater", so it should be the latest public everything. Can't do any more testing right now as I'm at work but later tonight I can try things out again.
I retested again , and basically just going to settings, and go through "Display/File requester/Advanced" cause a crash. I.e. you can run settings, adn the click on "Advanced" to have the same crash.
Each clicking on the Advanced tab runs a piece of code that checks the current available space in the temp directory. What is your setting for temporary files?
Each clicking on the Advanced tab runs a piece of code that checks the current available space in the temp directory. What is your setting for temporary files?
It was just plain download from os4depot / unpack / run , change nothing. But crash happens even if i just click on "File requester" tab too, right after spawn a settings.
Anyway, is this "temp" setting in rave's settings ? If so, i can only navigate in Display one, File Reuqest and Advanced one crashes
Hm, in that case it's something else. The temp directory defaults to PROGDIR:Temp/
When clicking on the File requester and Advanced listbrowser items, the program performs a WM_RETHINK. Perhaps your system has an old version of window.class? Just guessing.
Please let me know which version you have of: window.class layout.gadget listbrowser.gadget clicktab.gadget button.gadget string.gadget getfile.gadget
The Prefs window doesn't use any Enhancer gadgets, so I don't think the problem is there.
Same versions as yours, except for string.gadget - I have the new one that was released recently.
Looking at the code I can see that there's a SetGadgetAttrs() call before the WM_RETHINK, which triggers an unnecessary refresh. I've changed it to plain SetAttrs() because the rethink does the refresh automatically. But no idea whether it'll help fix the problem...
I have written a guest story for amiga-news.de's 2022 Advent Calendar, which you can read under this link. I'd like to thank Daniel Reimann of the Amiga-news Team for the opportunity!