Description
Describe the bug
There are two issues pointing to the same commit when I bisect:
- If I do a fresh import of raw or jpg files and then click on a file in lighttable to go to darkroom, then any action in darkroom (change rotation, change exposure) triggers a death skull icon in lighttable when you go back.
- If I import a TIFF, then that files gives me a death skull in lighttable right away
Steps to reproduce
- Compile git master from source and start darktable with a fresh config and cachedir.
- Import RAW (CR2, NEF and RW2), JPG and TIFF files.
- The TIFF files get death skulls in lighttable.
- Click on a raw file icon in lighttable (or jpg file).
- Execute any action, like change rotation or change exposure in exposure module.
- Go back to lighttable. The lighttable icon will now show a death skull.
Expected behavior
Darktable should display lighttable icons as usual when raw (or jpg) files are changes in darktable. TIFF-files should display normally as icons in lighttable after import.
Logfile | Screenshot | Screencast
Git bisect gives me this:
7dc0faf4652f8ed856702634634b90f4807dd6ec is the first bad commit
commit 7dc0faf4652f8ed856702634634b90f4807dd6ec
Author: wpferguson <[email protected]>
Date: Wed Sep 25 12:55:06 2024 -0400
fix dt_imageio_export_with_flags return values (#17529)
* src/imageio/imageio.c - changed dt_imageio_export_with_flags to return
TRUE on success and FALSE on error. Since
dt_imageio_export calls dt_imageio_export_with_flags
to do the actual export, dt_imageio_export now
returns TRUE for success and FALSE for error.
src/imageio/storage/disk.c
src/imageio/storage/email.c
src/imageio/storage/gallery.c
src/imageio/storage/pwigo.c
src/lua/luastorage.c - updated the places where dt_imageio_export was
called to use the boolean return value for
success tests.
src/imageio/imageio.c | 4 ++--
src/imageio/storage/disk.c | 4 ++--
src/imageio/storage/email.c | 4 ++--
src/imageio/storage/gallery.c | 8 ++++----
src/imageio/storage/piwigo.c | 4 ++--
src/lua/luastorage.c | 4 ++--
6 files changed, 14 insertions(+), 14 deletions(-)
Commit
7dc0faf is the first bad commit
Where did you obtain darktable from?
self compiled
darktable version
release-4.9.0-580-g3b57b3301e
What OS are you using?
Linux
What is the version of your OS?
Linux Debian 12.7
Describe your system?
- OS : Linux debian 6.1.0-25-amd64 1 SMP PREEMPT_DYNAMIC Debian 6.1.106-3 (2024-08-26) x86_64 GNU/Linux
- Linux - Distro : Debian 12.7 Stable Bookworm
- Memory : 64 GB
- Graphics card : NVIDIA Corporation TU104GL [Quadro RTX 4000]
- Graphics driver : NVIDIA 535.183.01 from Debian non-free repo
- OpenCL installed : Yes, OpenCL 3.0 CUDA 12.2.148
- Xorg : X11
- Desktop : Gnome 43.9
- GTK: 3.24.38-2~deb12u1, 4.8.3+ds-2+deb12u1, 2.24.33-2
- gcc : (GCC) 12.2.0
- cflags :
- CMAKE_BUILD_TYPE : Release
- Intel® Core™ i9-14900K × 32
- motherboard: ASUS PRIME Z790-A WIFI
Are you using OpenCL GPU in darktable?
Yes
If yes, what is the GPU card and driver?
NVIDIA Corporation TU104GL [Quadro RTX 4000], 535.183.01-1~deb12u1 amd64
Please provide additional context if applicable. You can attach files too, but might need to rename to .txt or .zip
**Can you reproduce with another darktable version(s)? 7dc0faf is the first bad commit
Can you reproduce with a RAW or Jpeg or both? raw, jpg and tiff behaves differently as stated in describe the bug. Raw and jpg import nicely to lighttable but get a death skull if manipulated in darkroom. Imported TIFF get a death skull in lighttable directly after import
Are the steps above reproducible with a fresh edit (i.e. after discarding history)? yes
If the issue is with the output image, attach an XMP file if (you'll have to change the extension to .txt)
Is the issue still present using an empty/new config-dir (e.g. start darktable with --configdir "/tmp")? yes
Do you use lua scripts? no
Open CL: Open CL on or off does not matter.