Bug688588.ps and Bug690160.ps (page 3) both have colorred/non-black pattern fills rendered partly as black going pxlcolor then pcl6 to ppm. Assigned to Marcos to try out on real PXL color printer to see whether it is pxlcolor or pcl6 (pcl6->pdf shows colored patterns) before sending them onto either me or Henry.
In both cases the issue appears to be with the pxlcolor device; sending the Ghostscript generated pxl file to an HP ColorLaserJet 4500 results in output that matches the GhostPCL output.
Thanks for the confirmation. Still a bit curious how pcl6->pdf looks different, but I have filed a few 'pcl6->pdf looks different from ppm' already, and it might just go away when those are fixed. I'll look properly why pattern fill is losing color in the middle - it appears that incomplete pattern tiles (those are at the edge) are colored correctly while complete ones (in the middle) are not.
Created attachment 5723 [details] paint through character set ROP to paint through character glyphs (pattern tiles). Apparently complete pattern tiles are sent as a character in a custom symbol font (rather than small rectangles for incomplete tiles), thus allowing re-use. The code set the fill color but never set ROP to "paint with brush color" thus glyphs appears as black. This patch set the ROP3 operation to paint through complete pattern tiles. This should also fix the bulk of the problems in bug 690585 (for ghostpdl).
patch was commited as r10403, and showed okay from Marcos before Christmas.