I've had a bug report opened that says that colord takes up more than 5% of the total time to boot on an embedded system. After doing some analysis with sysprof, colord spends the majority of it's time just generating the MD5 of the system profiles before it pushes the information on the bus.
Now, would I be right in saying that the pre-cooked md5 of the profile should be returned using the cmsGetHeaderProfileID? All the profiles on my system seem to return a profile ID of zero. Using iccdump, it looks like none of the profiles have any data for the profile id. Is this expected? If I remember correctly, the ProfileID was a ICC version 4 specification addition, and is also optional. Would it be sane to open the profiles on this embedded box, calculate the md5 using cmsMD5computeID, resave the profiles in the same location and then the problem would go away. The other alternative is to ship a smaller set of color profiles for this slow system. Ideas welcome. Thanks, Richard. ------------------------------------------------------------------------------ Xperia(TM) PLAY It's a major breakthrough. An authentic gaming smartphone on the nation's most reliable network. And it wants your games. http://p.sf.net/sfu/verizon-sfdev _______________________________________________ Lcms-user mailing list Lcms-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/lcms-user
