Summary: | emerging gnuplot on text console fails due to (sandbox) access vioation to /dev/mouse | ||
---|---|---|---|
Product: | Gentoo Linux | Reporter: | Zbigniew Koza <zkoza> |
Component: | New packages | Assignee: | Grant Goodyear (RETIRED) <g2boojum> |
Status: | RESOLVED FIXED | ||
Severity: | normal | CC: | andre.hinrichs, azarah, jakub, sci |
Priority: | High | ||
Version: | unspecified | ||
Hardware: | x86 | ||
OS: | Linux | ||
Whiteboard: | |||
Package list: | Runtime testing required: | --- | |
Attachments: | emerge log |
Description
Zbigniew Koza
2005-02-22 04:24:39 UTC
*** Bug 82941 has been marked as a duplicate of this bug. *** As stated in bug #80135, with udev /dev/mouse does not exist. It is /dev/input/mousen. So should gnuplot not be looking for the correct mouse device rather than /dev/mouse? I have a similar problem compiling gnuplot on a text console. But it's not the mouse device... When I merge that package remotely via ssh, it finishes without error! Very strange!! Here is the error: make[2]: Entering directory `/var/tmp/portage/gnuplot-4.0-r1/work/gnuplot-4.0.0/tutorial' svgalib: Cannot open /dev/svga Is svgalib_helper module loaded? make[2]: *** [eg1.tex] Error 1 make[2]: *** Waiting for unfinished jobs.... make[2]: Leaving directory `/var/tmp/portage/gnuplot-4.0-r1/work/gnuplot-4.0.0/tutorial' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/var/tmp/portage/gnuplot-4.0-r1/work/gnuplot-4.0.0' make: *** [all] Error 2 !!! ERROR: media-gfx/gnuplot-4.0-r1 failed. !!! Function src_compile, Line 67, Exitcode 2 !!! (no error message) !!! If you need support, post the topmost build error, NOT this status message. Andre: Try to (re-)emerge svgalib and than emerge gnuplot again. Probebly you don't have a svgalib_helper module in your kernel modules dir anymore (because you updated your kernel modules) Created attachment 55637 [details]
emerge log
After doing some more configurations on my system the svgalib_helper error
has gone. Unfortunately, now the described error appears here too.
See the attachment for the complete error message.
Just to play a bit I added the same hack as described in bug #41711 for /dev/svga to handle /dev/mouse and it compiles fine on console then! Means, just add "addwrite /dev/mouse" after the line "addwrite /dev/svga" Does this hack hurt the compile in other cases? I'm having the same trouble but the offending device is /dev/tts/0 Az, Better to add these addwrite's in the ebuild, or should they just be added to the list of globally-allowed targets in sandbox? Also, tossing to sci@gentoo.org, since I'm not maintaining gnuplot very well! Not sure if that is the best bet, as it might change for each user. Can anybody track exactly what tries to write to these devices ? Is ther any progress on fixing this? I'm still holding a local copy of the ebuild with the needed modification I described in comment #6. And I'm really surprised that so few people have problems with this. Why is that? Is it really that system dependent? I doubt, since I bought a new system four months ago and had the same problem with gnuplot. Or is anybody else compiling under X (which is fine)??? doing texcongif init as someone suggested on the forum didnt help. I seems it cant make pdf files or something with pdfetex ??? make[2]: Entering directory `/var/tmp/portage/gnuplot-4.0-r1/work/gnuplot-4.0.0/tutorial' Building LaTeX tutorial TEXINPUTS=.:.:: latex tutorial This is pdfeTeX, Version 3.141592-1.30.5-2.2 (Web2C 7.5.5) ---! /var/lib/texmf/web2c/latex.fmt doesn't match pdfetex.pool (Fatal format file error; I'm stymied) make[2]: *** [tutorial.dvi] Error 1 make[2]: Leaving directory `/var/tmp/portage/gnuplot-4.0-r1/work/gnuplot-4.0.0/tutorial' make[1]: *** [all-recursive] Error 1 make[1]: Leaving directory `/var/tmp/portage/gnuplot-4.0-r1/work/gnuplot-4.0.0' make: *** [all] Error 2 !!! ERROR: sci-visualization/gnuplot-4.0-r1 failed. Call stack: ebuild.sh, line 1933: Called dyn_compile ebuild.sh, line 971: Called src_compile this helped me: fmtutil-sys --all && fmtutil-sys --byfmt pdfetex (In reply to comment #11) (In reply to comment #12) Completely unrelated, move this elsewhere. --- BTW, what's the status here? Fixing the symptoms because we still don't know the cause. |