Siril version: 1.4.0-beta2 Note: tested with master branch, see below Operating System: Linux Package: Fedora RPMs (self packaged) # Description of the bug Siril crashes on stacking command: ``` $ siril WARNING: version mismatch between CFITSIO header (v40500) and linked library (v40600). Initializing FFTW multithreading support... log: Welcome to siril v1.4.0-beta2 log: Supported file types: BMP images, PIC images (IRIS), PGM and PPM binary images, RAW images, FITS-CFA images, Films, SER sequences, TIFF images, XISF images, JPG images, JPEG XL images, PNG images, HEIF images, AVIF images. log: Setting CWD (Current Working Directory) to '/mnt/partizione/foto/elaborazione/NGC7000/process' log: Parallel processing enabled: using 16 logical processors. log: Preparing python virtual environment: /home/mattia/.local/share/siril/venv. log: Color management active. log: JPEG ICC profiles supported. log: Checking the python module is up-to-date... log: Python module is up-to-date log: Running asnet version 0.97 log: Default FITS extension is set to .fit Entering wrapping_idle for function 0x558d619eecd0 log: Searching for scripts in: "/usr/share/siril/scripts"... log: Loading script: Mono_Preprocessing.ssf log: Loading script: OSC_Extract_Ha.ssf log: Loading script: OSC_Extract_HaOIII.ssf log: Loading script: OSC_Preprocessing.ssf log: Loading script: OSC_Preprocessing_BayerDrizzle.ssf log: Loading script: RGB_Composition.ssf log: Searching for scripts in: "/home/mattia/.siril/scripts"... log: Loading script: OSC_Preprocessing_BD_only.ssf log: Loading script: OSC_Preprocessing_BD_only_drizzle.ssf log: Loading script: OSC_Preprocessing_WithoutDark.ssf log: Loading script: OSC_Preprocessing_WithoutDBF_drizzle.ssf log: Loading script: OSC_Preprocessing_WithoutDBF_new.ssf Entering wrapping_idle for function 0x558d619c33d0 log: Gaia archive available Last available version: 1.4.0-beta2 input command:stack r_pp_light rej 3 3 -norm=addscale -output_norm -rgb_equal -out=result log: Running command: stack 1748764874: running command stack Reading sequence file `r_pp_light.seq'. Reading sequence failed, file cannot be opened: r_pp_light.seq. Reading sequence file `r_pp_light_.seq'. bitpix for the sequence is set as -32 Reading sequence file `r_pp_light_.seq'. bitpix for the sequence is set as -32 log: Stacking sequence r_pp_light_ number of filtered-in images: 20 number of filtered-in images: 20 log: Processing all images in the sequence (20) log: Stacking result will be stored as a 32-bit image log: Computing normalization... Memory per image: 462 MB. Max memory: 18211 MB log: With the current memory and thread (16) limits, up to 16 thread(s) can be used for sequence normalization Writing sequence file r_pp_light_.seq log: Normalization computation time: 10.13 ms image size: 6024x4022, 3 layers cfitsio was compiled with multi-thread support, stacking will be executed by several cores log: Using 17934 MB memory maximum for stacking log: We have 30 parallel blocks of size 402 (+2) for stacking. Block 0: channel 0, from 0 to 402 (h = 403) Block 1: channel 0, from 403 to 805 (h = 403) Block 2: channel 0, from 806 to 1207 (h = 402) Block 3: channel 0, from 1208 to 1609 (h = 402) Block 4: channel 0, from 1610 to 2011 (h = 402) Block 5: channel 0, from 2012 to 2413 (h = 402) Block 6: channel 0, from 2414 to 2815 (h = 402) Block 7: channel 0, from 2816 to 3217 (h = 402) Block 8: channel 0, from 3218 to 3619 (h = 402) Block 9: channel 0, from 3620 to 4021 (h = 402) Block 10: channel 1, from 0 to 402 (h = 403) Block 11: channel 1, from 403 to 805 (h = 403) Block 12: channel 1, from 806 to 1207 (h = 402) Block 13: channel 1, from 1208 to 1609 (h = 402) Block 14: channel 1, from 1610 to 2011 (h = 402) Block 15: channel 1, from 2012 to 2413 (h = 402) Block 16: channel 1, from 2414 to 2815 (h = 402) Block 17: channel 1, from 2816 to 3217 (h = 402) Block 18: channel 1, from 3218 to 3619 (h = 402) Block 19: channel 1, from 3620 to 4021 (h = 402) Block 20: channel 2, from 0 to 402 (h = 403) Block 21: channel 2, from 403 to 805 (h = 403) Block 22: channel 2, from 806 to 1207 (h = 402) Block 23: channel 2, from 1208 to 1609 (h = 402) Block 24: channel 2, from 1610 to 2011 (h = 402) Block 25: channel 2, from 2012 to 2413 (h = 402) Block 26: channel 2, from 2414 to 2815 (h = 402) Block 27: channel 2, from 2816 to 3217 (h = 402) Block 28: channel 2, from 3218 to 3619 (h = 402) Block 29: channel 2, from 3620 to 4021 (h = 402) allocating data for 16 threads (each 185 MB) log: Starting stacking... log: Pixel rejection in channel #0: 0.459% - 0.784% log: Pixel rejection in channel #1: 0.320% - 0.727% log: Pixel rejection in channel #2: 0.368% - 0.683% free and close (0) log: Rejection stacking complete. 20 images have been stacked. log: Integration of 20 images on 20 of the sequence: log: Pixel combination ......... average log: Input normalization ....... additive + scaling log: Output normalization ...... enabled log: Pixel rejection ........... winsorized sigma clipping log: Rejection parameters ...... low=3.000 high=3.000 log: Creating rejection maps ... no log: Image weighting ........... disabled log: RGB equalization .......... enabled (siril:13400): GLib-ERROR **: 10:01:20.044: file ../glib/gthread-posix.c: line 796 (g_system_thread_wait): error 'Evitato uno stallo di risorse (deadlock)' during 'pthread_join (pt->system_thread, NULL)' Error, signal 5: siril(+0xcf580) [0x558d61874580] /lib64/libc.so.6(+0x19c30) [0x7fc314627c30] /lib64/libglib-2.0.so.0(g_log_structured_array+0x123) [0x7fc31aa37d53] /lib64/libglib-2.0.so.0(g_log_default_handler+0xd7) [0x7fc31aa37e47] /lib64/libglib-2.0.so.0(g_logv+0x247) [0x7fc31aa32477] /lib64/libglib-2.0.so.0(g_log+0x93) [0x7fc31aa32803] /lib64/libglib-2.0.so.0(g_thread_join+0xcd) [0x7fc31aa5bf8d] siril(stop_processing_thread+0x73) [0x558d6186bc83] siril(+0xac4cf) [0x558d618514cf] siril(+0xadfe8) [0x558d61852fe8] /lib64/libglib-2.0.so.0(+0x76862) [0x7fc31aa62862] /lib64/libc.so.6(+0x711d4) [0x7fc31467f1d4] /lib64/libc.so.6(+0xf3cec) [0x7fc314701cec] ``` With 1.4.0-beta1 the same stacking command on the same sources is performed well. With 1.4.0-beta2 the above crash happens. With master branch as of today (commit a12a95474ce5074c2953eff925e8a1e40e92343d) it doesn't crash, but instead it stays stuck on "Finalizing stacking" forever: ``` $ siril WARNING: version mismatch between CFITSIO header (v40500) and linked library (v40600). Initializing FFTW multithreading support... log: Welcome to siril v1.4.0-beta3 log: Supported file types: BMP images, PIC images (IRIS), PGM and PPM binary images, RAW images, FITS-CFA images, Films, SER sequences, TIFF images, XISF images, JPG images, JPEG XL images, PNG images, HEIF images, AVIF images. log: Setting CWD (Current Working Directory) to '/mnt/partizione/foto/elaborazione/NGC7000/process' log: Parallel processing enabled: using 16 logical processors. log: Preparing python virtual environment: /home/mattia/.local/share/siril/venv. log: Color management active. log: JPEG ICC profiles supported. log: Checking the python module is up-to-date... log: Python module is up-to-date log: Running asnet version 0.97 log: Default FITS extension is set to .fit Entering wrapping_idle for function 0x5646e73e5ae0 log: Searching for scripts in: "/usr/share/siril/scripts"... log: Loading script: Mono_Preprocessing.ssf log: Loading script: OSC_Extract_Ha.ssf log: Loading script: OSC_Extract_HaOIII.ssf log: Loading script: OSC_Preprocessing.ssf log: Loading script: OSC_Preprocessing_BayerDrizzle.ssf log: Loading script: RGB_Composition.ssf log: Searching for scripts in: "/home/mattia/.siril/scripts"... log: Loading script: OSC_Preprocessing_BD_only.ssf log: Loading script: OSC_Preprocessing_BD_only_drizzle.ssf log: Loading script: OSC_Preprocessing_WithoutDark.ssf log: Loading script: OSC_Preprocessing_WithoutDBF_drizzle.ssf log: Loading script: OSC_Preprocessing_WithoutDBF_new.ssf Entering wrapping_idle for function 0x5646e73ba250 Last available version: 1.4.0-beta2 log: Gaia archive available input command:stack r_pp_light rej 3 3 -norm=addscale -output_norm -rgb_equal -out=result log: Running command: stack 1748766664: running command stack Reading sequence file `r_pp_light.seq'. Reading sequence failed, file cannot be opened: r_pp_light.seq. Reading sequence file `r_pp_light_.seq'. bitpix for the sequence is set as -32 Reading sequence file `r_pp_light_.seq'. bitpix for the sequence is set as -32 log: Stacking sequence r_pp_light_ number of filtered-in images: 20 number of filtered-in images: 20 log: Processing all images in the sequence (20) log: Stacking result will be stored as a 32-bit image log: Computing normalization... Memory per image: 462 MB. Max memory: 12267 MB log: With the current memory and thread (16) limits, up to 16 thread(s) can be used for sequence normalization Writing sequence file r_pp_light_.seq log: Normalization computation time: 15.41 ms image size: 6024x4022, 3 layers cfitsio was compiled with multi-thread support, stacking will be executed by several cores log: Using 11990 MB memory maximum for stacking log: We have 30 parallel blocks of size 402 (+2) for stacking. Block 0: channel 0, from 0 to 402 (h = 403) Block 1: channel 0, from 403 to 805 (h = 403) Block 2: channel 0, from 806 to 1207 (h = 402) Block 3: channel 0, from 1208 to 1609 (h = 402) Block 4: channel 0, from 1610 to 2011 (h = 402) Block 5: channel 0, from 2012 to 2413 (h = 402) Block 6: channel 0, from 2414 to 2815 (h = 402) Block 7: channel 0, from 2816 to 3217 (h = 402) Block 8: channel 0, from 3218 to 3619 (h = 402) Block 9: channel 0, from 3620 to 4021 (h = 402) Block 10: channel 1, from 0 to 402 (h = 403) Block 11: channel 1, from 403 to 805 (h = 403) Block 12: channel 1, from 806 to 1207 (h = 402) Block 13: channel 1, from 1208 to 1609 (h = 402) Block 14: channel 1, from 1610 to 2011 (h = 402) Block 15: channel 1, from 2012 to 2413 (h = 402) Block 16: channel 1, from 2414 to 2815 (h = 402) Block 17: channel 1, from 2816 to 3217 (h = 402) Block 18: channel 1, from 3218 to 3619 (h = 402) Block 19: channel 1, from 3620 to 4021 (h = 402) Block 20: channel 2, from 0 to 402 (h = 403) Block 21: channel 2, from 403 to 805 (h = 403) Block 22: channel 2, from 806 to 1207 (h = 402) Block 23: channel 2, from 1208 to 1609 (h = 402) Block 24: channel 2, from 1610 to 2011 (h = 402) Block 25: channel 2, from 2012 to 2413 (h = 402) Block 26: channel 2, from 2414 to 2815 (h = 402) Block 27: channel 2, from 2816 to 3217 (h = 402) Block 28: channel 2, from 3218 to 3619 (h = 402) Block 29: channel 2, from 3620 to 4021 (h = 402) allocating data for 16 threads (each 185 MB) log: Starting stacking... log: Pixel rejection in channel #0: 0.459% - 0.784% log: Pixel rejection in channel #1: 0.320% - 0.727% log: Pixel rejection in channel #2: 0.368% - 0.683% free and close (0) log: Rejection stacking complete. 20 images have been stacked. log: Integration of 20 images on 20 of the sequence: log: Pixel combination ......... average log: Input normalization ....... additive + scaling log: Output normalization ...... enabled log: Pixel rejection ........... winsorized sigma clipping log: Rejection parameters ...... low=3.000 high=3.000 log: Creating rejection maps ... no log: Image weighting ........... disabled log: RGB equalization .......... enabled Entering wrapping_idle for function 0x5646e7268f70 ```