| Summary: | Non-deterministic tiffsep output under certain circumstances | ||
|---|---|---|---|
| Product: | Ghostscript | Reporter: | Jason Giglio <gigs> |
| Component: | Other Driver | Assignee: | Michael Vrhel <michael.vrhel> |
| Status: | RESOLVED FIXED | ||
| Severity: | normal | CC: | alex |
| Priority: | P4 | ||
| Version: | master | ||
| Hardware: | PC | ||
| OS: | Linux | ||
| Customer: | Word Size: | --- | |
Created attachment 9083 [details]
Test PDF 2
Valgrind reports many warnings starting with: Conditional jump or move depends on uninitialised value(s) gsicc_get_link (gsicc_cache.c:590) gx_concretize_ICC (gsicc.c:524) gx_concretize_Separation (gscsepr.c:389) gx_default_remap_color (gxcmap.c:600) gx_remap_Separation (gscsepr.c:302) capture_spot_equivalent_cmyk_colors (gsequivc.c:515) update_Separation_spot_equivalent_cmyk_colors (gsequivc.c:145) update_spot_equivalent_cmyk_colors (gsequivc.c:311) tiffsep_update_spot_equivalent_colors (gdevtsep.c:850) gx_install_Separation (gscsepr.c:136) gs_setcolorspace_only (gscolor2.c:46) gs_setcolorspace (gscolor2.c:64) This no longer shows any valgrind issues. |
Created attachment 9082 [details] Test PDF 1 When outputting the attached files which are the same other than colorant naming, sometimes there are regressions in the outputted tiffs and sometimes there are not. It appears to be a non-deterministic bug of some sort. gsc -dNOPAUSE -dBATCH -sDEVICE=tiffsep -sOutputFile=/tmp/test1/files -f original.pdf gsc -dNOPAUSE -dBATCH -sDEVICE=tiffsep -sOutputFile=/tmp/test2/files -f renamed.pdf Then for each separation and the RGB: tiffcmp /tmp/test1/files /tmp/test2/files | grep -v "DateTime"