Bug: "working..." stuck on screen and module(s) have stopped

[Annabel2]

Code of Conduct

• I have read this document and I accept the conditions

Search previous report

• I have searched my issue

Computer's operating system

Linux

What GPU do you have ?

No GPU

Ansel version

0.0.0+2009~gca34cbc24

Type of binary

Pre-built packages provided by the Ansel project (.AppImage, .EXE)

What happened?

I did some pretty routine editing then used Diffuse/Sharpen to do a hard deblur. I think it did it but the message "working..." has remained on screen, still there as I type this. It appears I can carry on editing pretty much. However if I zoom in to 100%, the display is initially a bit blurred as usual, but then it never sharpens up. "working.." is there all the time. The lack of sharpening up in confirmed by the system monitor not showing the usual splurge of CPU activity. I switch D/S off. no CPU. I zoom to "fit screen", no CPU. There's nothing weird in Terminal, lots of ROI messages.

Does your problem occur with OpenCL enabled ?

I don't know

Does your problem occur with OpenCL disabled ?

Yes

Previous Ansel version that use to work

I use build 1975 mainly, this does not have this issue.

Output of ansel -d all

Backtrace

[Jiyone]

It looks like the full pipe doesn't compute.

I advise you to use the 0.0+1975 version of Ansel before I finish working on scaling.

[Annabel2]

Thanks.
I'm out of touch with what's happening with Ansel. You're obviously working on it, that's good. Is Aurelien active on Ansel? It's all been quite quiet for a while.

[Jiyone]

Aurélien is not working anymore on Ansel for personal reasons.

[sahlirc]

Thank you for keeping this project alive!

[Annabel2]

@Jiyone , re. Aurelien that's a great shame, sending him best wishes.
As you probably know, links about the project that used to work no longer work. I'm up for doing some testing if that helps. I'd need to know what to test of course! So I think you'd need to provide me with some info about new builds as they become available, and what needs testing in a new build. It'd need to be fairly informal and not too technical!

[Jiyone]

What links are you talking about ?
You can see what's new here : https://github.com/aurelienpierreeng/ansel/commits/master/
For the moment there is nothing to test as nothing major has been committed.

[Annabel2]

This one (bookmarked re. "what changes are in an appimage") - https://community.ansel.photos/view-discussion/what-changes-are-in-an-appimage
?the community stuff got hacked?
Thanks for your link.

[Jiyone]

Yes the community site is down, but the rest of the site is working https://ansel.photos

[Jiyone]

Anyway, @Annabel2 can you reproduce ? If yes, how to ? Maybe run ansel with -d perf could show which module in the pipe Ansel "freezes".

[Annabel2]

Just reproduced this no problem. Here is the last section of Terminal -

rawprepare: ROI in: 8896x5920, ROI out: 8726x5840
temperature: ROI in: 8726x5840, ROI out: 8726x5840
highlights: ROI in: 8726x5840, ROI out: 8726x5840
demosaic: ROI in: 8726x5840, ROI out: 8726x5840
cacorrectrgb: ROI in: 8726x5840, ROI out: 8726x5840
initialscale: ROI in: 8726x5840, ROI out: 1675x1121
exposure: ROI in: 1675x1121, ROI out: 1675x1121
exposure: ROI in: 1675x1121, ROI out: 1675x1121
diffuse: ROI in: 1675x1121, ROI out: 1675x1121
colorin: ROI in: 1675x1121, ROI out: 1675x1121
channelmixerrgb: ROI in: 1675x1121, ROI out: 1675x1121
diffuse: ROI in: 1675x1121, ROI out: 1675x1121
defringe: ROI in: 1675x1121, ROI out: 1675x1121
atrous: ROI in: 1675x1121, ROI out: 1675x1121
colorbalancergb: ROI in: 1675x1121, ROI out: 1675x1121
colorbalancergb: ROI in: 1675x1121, ROI out: 1675x1121
filmicrgb: ROI in: 1675x1121, ROI out: 1675x1121
tonecurve: ROI in: 1675x1121, ROI out: 1675x1121
colorout: ROI in: 1675x1121, ROI out: 1675x1121
gamma: ROI in: 1675x1121, ROI out: 1675x1121
image colorspace transform Lab-->RGB took 0.007 secs (0.040 CPU) [diffuse ]
284.427587 [dt_dev_pixelpipe_change] pipeline resync on the current modules stack took 0.013 secs (0.059 CPU) for pipe 4
rawprepare: ROI in: 8896x5920, ROI out: 8726x5840
temperature: ROI in: 8726x5840, ROI out: 8726x5840
highlights: ROI in: 8726x5840, ROI out: 8726x5840
demosaic: ROI in: 8726x5840, ROI out: 8726x5840
cacorrectrgb: ROI in: 8726x5840, ROI out: 8726x5840
initialscale: ROI in: 8726x5840, ROI out: 1675x1121
exposure: ROI in: 1675x1121, ROI out: 1675x1121
exposure: ROI in: 1675x1121, ROI out: 1675x1121
diffuse: ROI in: 1675x1121, ROI out: 1675x1121
colorin: ROI in: 1675x1121, ROI out: 1675x1121
channelmixerrgb: ROI in: 1675x1121, ROI out: 1675x1121
diffuse: ROI in: 1675x1121, ROI out: 1675x1121
defringe: ROI in: 1675x1121, ROI out: 1675x1121
atrous: ROI in: 1675x1121, ROI out: 1675x1121
colorbalancergb: ROI in: 1675x1121, ROI out: 1675x1121
colorbalancergb: ROI in: 1675x1121, ROI out: 1675x1121
filmicrgb: ROI in: 1675x1121, ROI out: 1675x1121
tonecurve: ROI in: 1675x1121, ROI out: 1675x1121
colorout: ROI in: 1675x1121, ROI out: 1675x1121
gamma: ROI in: 1675x1121, ROI out: 1675x1121
284.483545 [dev_pixelpipe] took 0.000 secs (0.000 CPU) initing base buffer [preview]
284.544823 [dev_pixelpipe] took 0.126 secs (0.607 CPU) processed diffuse or sharpen' on CPU, blended on CPU [full] image colorspace transform RGB-->Lab took 0.003 secs (0.025 CPU) [defringe 0] 284.588948 [dev_pixelpipe] took 0.044 secs (0.242 CPU) processed defringe' on CPU, blended on CPU [full]
284.803909 [dev_pixelpipe] took 0.215 secs (1.449 CPU) processed contrast equalizer' on CPU, blended on CPU [full] image colorspace transform Lab-->RGB took 0.006 secs (0.034 CPU) [colorbalancergb ] 284.993941 [dev_pixelpipe] took 0.190 secs (1.326 CPU) processed color balance rgb' on CPU, blended on CPU [full]
[raster masks] mask id 0 from exposure (darks) for module color balance rgb (darks) could not be found in pipe 2. Pipe re-entry will be attempted.
285.188573 [dev_pixelpipe] took 0.195 secs (1.303 CPU) processed `color balance rgb darks' on CPU, blended on CPU [full]
306.640620 [dt_dev_history_auto_save] auto-saving history upon last change took 0.037 secs (0.026 CPU)

This is the XMP just before I went back into the "darktable" and attempted a hard deblur

_5D_05875.CR2-copy.xmp.txt

Here is the XMP with Ansel hung and showing "working..."

_5D_05875.CR2.xmp.txt

[lukadh]

I had a quick look at the XMPs. Here are the differences (before and after):

[...]
darktable:change_timestamp="63900744867962991"
darktable:change_timestamp="63900744951199034"

[...]

darktable:history_end="27"
darktable:history_end="28"

[...]

darktable:history_current_hash="1e117291dff2c2b662745ccc008548bb">
darktable:history_current_hash="4bc80a8e91c53988a6db542b4b82dbd5">

[...]

[blank]
<rdf:li
      darktable:num="27"
      darktable:operation="diffuse"
      darktable:enabled="1"
      darktable:modversion="2"
      darktable:params="18000000000000000c000000000040400000803f0000803f000000000000803f0000000000000000000080be0000003e000000bf0000803e00000000"
      darktable:multi_name=""
      darktable:multi_priority="0"
      darktable:blendop_version="11"
      darktable:blendop_params="gz10eJxjYGBgYAFiCQYYOOHEgAZY0QVwggZ7CB6pfOygYtaVAyCMi08IAAB/xiOk"/>