Summary: | media-gfx/digikam-3.5.0: image editor tools slow with media-libs/lcms-2.5-r1 (colormanagement) | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Peter Albrecht <px79> |
Component: | [OLD] KDE | Assignee: | Andreas K. Hüttel <dilfridge> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | kde |
Priority: | Normal | ||
Version: | unspecified | ||
Hardware: | AMD64 | ||
OS: | Linux | ||
URL: | https://bugs.kde.org/show_bug.cgi?id=322789 | ||
Whiteboard: | fixed in 4.x | ||
Package list: | Runtime testing required: | --- | |
Attachments: | compressed strace log file, see comment #2 |
Description
Peter Albrecht
2013-12-27 09:34:43 UTC
No idea yet, I can't reproduce the problem... Is media-libs/lcms-1* still installed on your system? If yes, could you please remove it? emerge -ac media-libs/lcms:0 emerge @preserved-rebuild Does that help? If you can't unmerge lcms:0, what is still pulling it in? Additional debugging idea, emerge dev-util/strace and then do the following: 1) make sure a clock with seconds display is visible on your screen (e.g., kde bar) 2) in a terminal window start digikam with strace -tt digikam &> digikam-strace.log (note, because of the logging digikam will be slower in any case) 3) note exact time, start one of the "unusually slow tasks" 4) close digikam then take the log file (which will be HUGE) and shorten it so only the parts around that particular time are there, compress it and upload it (warning: it will contain things like file names and file paths...) maybe we can see something from the log (In reply to Andreas K. Hüttel from comment #1) > Is media-libs/lcms-1* still installed on your system? If yes, could you > please remove it? > emerge -ac media-libs/lcms:0 > emerge @preserved-rebuild > Does that help? > > If you can't unmerge lcms:0, what is still pulling it in? _Short_: Did not help. _Long_: Yes, media-libs/lcms-1.19 was still installed on my system. I removed it the following way: 1) change use-flags of some programms to "-lcms" (e.g. gimp) 2) emerge --update --newuse --deep world 3) emerge -va @preserved-rebuild 4) emerge -C media-gfx/inkscape-0.48.4 5) emerge --depclean (removed lcms-1.19) 6) revdep-rebuild 7) dispatch-conf Now "eix media-libs/lcms" tells me, "2.5-r1(2)" is the only installed version of lcms. I rebuild digiKam and restarted the system. But the initial bug (long waiting time) is still present. Created attachment 366330 [details] compressed strace log file, see comment #2 (In reply to Andreas K. Hüttel from comment #2) > Additional debugging idea, emerge dev-util/strace and then do the following: > [...] > 2) in a terminal window start digikam with > strace -tt digikam &> digikam-strace.log > (note, because of the logging digikam will be slower in any case) Andreas, this was a very good idea! I created the strace log file, as described and uploaded it to this bug. Very interesting is the fact that you can find 464 times the following lines: ------------------------------------- 8< ------------------------------------- open("/usr/share/apps/libkdcraw/profiles/srgb-d65.icm", O_RDONLY|O_CLOEXEC) = 65 fcntl(65, F_SETFD, FD_CLOEXEC) = 0 fstat(65, {st_mode=S_IFREG|0644, st_size=6924, ...}) = 0 fstat(65, {st_mode=S_IFREG|0644, st_size=6924, ...}) = 0 fstat(65, {st_mode=S_IFREG|0644, st_size=6924, ...}) = 0 read(65, "\0\0\33\flcms\0020\0\0mntrRGB XYZ \7\324\0\10\0\r\0\f"..., 16384) = 6924 read(65, "", 9460) = 0 close(65) = 0 ------------------------------------- >8 ------------------------------------- DigiKam seems to read this profile file 464 times. 6924 is the exact size in bytes of this file on my harddisc. But why does digiKam read this file so often? (In reply to Peter Albrecht from comment #5) > DigiKam seems to read this profile file 464 times. 6924 is the exact size in > bytes of this file on my harddisc. But why does digiKam read this file so > often? Yeah I've seen that too... no idea why, but we can ask Gilles, maybe he has an idea. As a doublecheck you could rename that file as root and see what happens. (Possibly some error though.) I just tested this bug with digiKam 4.2.0 and it seems to have been fixed. :) https://bugs.kde.org/show_bug.cgi?id=322789#c41 (In reply to Peter Albrecht from comment #7) > I just tested this bug with digiKam 4.2.0 and it seems to have been fixed. :) > > https://bugs.kde.org/show_bug.cgi?id=322789#c41 Thanks. |