When I boot with splashutils during * setting user font.... someting fails. Reproducible: Always Steps to Reproduce: 1.emerge media-gfx/splashutils-0.9_rc1 2.configure the package 3.put CONSOLEFONT="lat0-16" in /etc/rc.conf 4.reboot Actual Results: if splash=silent after starting /etc/init.d/consolefont it switch to splash=verbose but not refresh the fb image correctly and printout "putfont: PIO_FONT: Invalid argument" if splash=verbose after consolefont the fb image shitch to strange color, to restore the right color switch to vc/2 and after to vc/1 Expected Results: splash=silent should not crash after consolefont switching from silent to verbose should refresh the fb image splash=verbose should not currupt colors after consolefont I'm using radeonfb on iBook G4 1Ghz yaboot option: image=/boot/kernel-2.6.9-gentoo-r9-sleep label=splash root=/dev/hda3 initrd=/boot/initrd-1024x768 read-only append= "video=radeonfb:1024x768-32 splash=silent,theme:nolith" emerge info: Portage 2.0.51-r3 (default-ppc-2004.3, gcc-3.4.1, glibc-2.3.3.20040420-r2, 2.6.9-gentoo-r9-sleep ppc) ================================================================= System uname: 2.6.9-gentoo-r9-sleep ppc 7447A, altivec supported Gentoo Base System version 1.4.16 Autoconf: sys-devel/autoconf-2.59-r5 Automake: sys-devel/automake-1.8.5-r1 Binutils: sys-devel/binutils-2.15.90.0.3-r3 Headers: sys-kernel/linux26-headers-2.6.8.1 Libtools: sys-devel/libtool-1.5.2-r7 ACCEPT_KEYWORDS="ppc" AUTOCLEAN="yes" CFLAGS="-Os -mcpu=G4 -maltivec -mabi=altivec -mpowerpc-gfxopt -fomit-frame-pointer -fno-strict-aliasing -pipe" CHOST="powerpc-unknown-linux-gnu" COMPILER="" CONFIG_PROTECT="/etc /usr/X11R6/lib/X11/xkb /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/mozilla/defaults/pref /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control"CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-Os -mcpu=G4 -maltivec -mabi=altivec -mpowerpc-gfxopt -fomit-frame-pointer -fno-strict-aliasing -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs ccache distlocks" GENTOO_MIRRORS="ftp://pandemonium.tiscali.de/pub/gentoo/ http://gentoo.mirror.icd.hu/ ftp://sunsite.informatik.rwth-aachen.de/pub/Linux/gentoo" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="" SYNC="rsync://rsync.gentoo.org/gentoo-portage" USE="S3TC X aac alsa altivec bash-completion berkdb bitmap-fonts bluetooth bootsplash cdparanoia cdr codecs crypt cscope css cups divx4linux dmx dts dvd dvdread encode esd ethereal fbcon flac font-server foomaticdb fortran gdbm gif gimp gimpprint gmail gnome gnome-libs gnomedb gnutls gpm gtk gtk2 hal icq ieee1394 imap imlib jabber java javascript jpeg lcd libwww mad mime mono mozilla moznoirc moznomail mozsvg mozxmlterm mp3 mpeg mplayer msn nas ncurses nls nntp no-old-linux nowin offensive oggvorbis opengl oss pam perl png ppc ppds python readline samba scanner sdl snmp socks5 softmmu spell sqlite ssl svg tcpd tetex theora tools truetype unicode usb v4l v4l2 vim vim-with-x vnc xanim xine xscreensaver xvid yahoo zlib"
same problem here.
This seems a duplicate of two previously closed bugs: Bug 63137 - gensplash switches to verbose of setting user font Bug 77121 - colour corruption; fb background not correctly painted upon switching to verbose.
Oops, Bug 77121 should read Bug 71121 :)
the same problem happens to me here too. I am using radeonfb as well. It works OK on 32 bits except for white lines or stokes of disturbance usually when things are drawing, e.g. progress bar The problem is indeed that when error shows up on consolefont splash drops from silent to verbose. happens both with splashutils-0.9_rc1 and splashutils-0.9_pre10. Spock, another problem I discovered on my system and told you already (this is a reminder... :) is that I have my kernel set with 1280x800-32 and fbset -i shows a geometry of 1280 800 1280 1024 32. 1024 here probably represents virtual res right? so why, where does it come from, i haven't set anything to be virtual resolution, so why should it be? Thanks
Alessio: please upgrade to a 2.6.10 g-d-s kernel. Claessens: are you using 2.6.10? If not, please upgrade. Gad: Could you please describe the distortions you see in more detail? Can you reproduce some of them by simply running `fbtruetype "some text"`? Or maybe by running `splash_util -c paint -t emergence -p 30000 -m s`? If you can, please take a shot of the framebuffer (with fbgrab, or simple `cat /dev/fb0 > some_file`) and send it to me/attach it to the bug. The virtual res is actually OK. It's usually up to the driver to choose a virtual resolution. If it's 1024 in your case, it is so because that size was chosen by radeonfb.
thanks for replying. I don't know how to discribe the disturbance. It is some very fine line, very small, from left to right on the screen, randomly showing across the screen and disappearing immediately. they last a tiny short moment, and it's impossible to take a screenshot of that. (maybe a video). The splash_util paint occasionally does cause the disturbance to happen, but it's much too short and small. the setpic command of splash_util for example would show a much bigger effect of disturbance. About the virtual resolution: while splash works it doesn't bother at all, but if I for example boot a kernel that does not have fbsplash but does have the framebuffer with that resolution, i won't be able to see the bottom part of the console, thus can't see where the cursor is or what is it i'm typing. It shouldn't be so IMO, but since it does not bother fbsplash, it's not a major problem.
with 2.6.10 g-d-s there's no disturbance and switching from silent ro verbose works well as aspected I'm waiting for iBook sleeeping patch for 2.6.10 source
Hi! Serveral other people experience the same problem: http://forums.gentoo.org/viewtopic.php?t=138391 This thread says that our problem is a conflict between setfont and fbsplash (with diff to 2.6.9 kernel). Some other solutions are also posted. Here's what happens on my system: If the splashscreen is in silent mode and /etc/init.d/consolefont is started, the screen switches to verbose mode (but in contrast to Alessio there is no problem with corrupted messages) and prints "putfont: PIO_FONT: Invalid argument". After my systems has booted up completely my latin font for displaying umlauts on the console is not working (it seems the default font is used). If I switch into verbose mode before /etc/init.d/consolefont is started, I don't get the putfont error and my latin font is loaded correctly. I don't have got the problems with strange colors as reported by Alessio. My graphics card is a 8MB Matrox Millenium G200 AGP. Portage 2.0.51-r15 (default-linux/x86/2004.3, gcc-3.3.5, glibc-2.3.4.20040808-r1, 2.6.10-gentoo-r6 i686) ================================================================= System uname: 2.6.10-gentoo-r6 i686 AMD Duron(tm) Gentoo Base System version 1.4.16 Python: dev-lang/python-2.3.4 [2.3.4 (#1, Nov 27 2004, 10:44:33)] dev-lang/python: 2.3.4 sys-devel/autoconf: 2.13, 2.59-r6 sys-devel/automake: 1.5, 1.8.5-r2, 1.6.3, 1.7.9, 1.4_p6, 1.9.4 sys-devel/binutils: 2.15.92.0.2-r1 sys-devel/libtool: 1.5.10-r3 virtual/os-headers: 2.4.21-r1 ACCEPT_KEYWORDS="x86" AUTOCLEAN="yes" CFLAGS="-O2 -march=athlon -fomit-frame-pointer -pipe" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/kde/2/share/config /usr/kde/3/share/config /usr/lib/X11/xkb /usr/share/config /usr/share/texmf/dvipdfm/config/ /usr/share/texmf/dvips/config/ /usr/share/texmf/tex/generic/config/ /usr/share/texmf/tex/platex/config/ /usr/share/texmf/xdvi/ /var/qmail/control" CONFIG_PROTECT_MASK="/etc/gconf /etc/terminfo /etc/env.d" CXXFLAGS="-O2 -march=athlon -fomit-frame-pointer -pipe" DISTDIR="/usr/portage/distfiles" FEATURES="autoaddcvs autoconfig ccache distlocks sandbox sfperms" GENTOO_MIRRORS="ftp://sunsite.informatik.rwth-aachen.de/pub/Linux/gentoo ftp://ftp.gentoo.mesh-solutions.com/gentoo/ ftp://linux.rz.ruhr-uni-bochum.de/gentoo-mirror/" MAKEOPTS="-j2" PKGDIR="/usr/portage/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/usr/local/portage" SYNC="rsync://rsync.europe.gentoo.org/gentoo-portage" USE="x86 3dnow X alsa apm avi berkdb bitmap-fonts crypt divx4linux dvd emacs encode esd f77 fam font-server fortran gdbm gif gnome gtk gtk2 imlib java jpeg libg++ libwww mad mikmod mmx motif mpeg ncurses oggvorbis opengl pam pdflib perl png python qt quicktime readline sdl spell ssl svga tcpd tetex tiff truetype truetype-fonts type1-fonts xml2 xmms xv zlib" Unset: ASFLAGS, CBUILD, CTARGET, LDFLAGS Best regards Christian
Spock, I just saw that you fixed the bug... Many thanks! http://dev.gentoo.org/~spock/projects/gensplash/#changelog
Yup, the problem should be fixed in splashutils-0.9.1. Closing the bug.
The PUTFONT bug does not seem to be resolved. After upgrading to the newest splasutils 0.9.1 (which is still masked as ~x86 at this time), the error is still displayed. So from a user point of view nothing has changed from the last to the current version. There is also a second user reporting the same problem in the forums: http://forums.gentoo.org/viewtopic.php?t=138391&start=25#2034678
The PUTFONT message is still there, because you just can't set fonts on the console when it's in the KD_GRAPHICS mode (and it is in that mode when fbsplash is running in the silent mode). I can do nothing about that (apart from changing the code in the kernel, which I'm not going to do because that would be a modification of the default behaviour). However, with the latest splashutils version, consolefont is run silently at the end of the boot process - that way you get the font you have selected in your rc configs. Have you experienced any problems with console fonts with the latest version of splashutils? (apart from just seeing the PUTFONT message, that is)
> Have you experienced any problems with console fonts with the latest version of splashutils? No. The font I have selected in rc.conf is now loaded correctly. > However, with the latest splashutils version, consolefont is run silently at the end of the boot process - that way you get the font you have selected in your rc configs. OK. If I understand you correctly, consolefont runs twice: * The first time when it displays the PUTFONT error * The second time when it is run silently ==> So I disabled consolefont via rc-update. As a result, the PUTFONT error is not shown anymore, but the splash switches to verbosemode around the point where runlevel 3 is entered (without error msg). I removed keymaps via rc-update because I thought this script could have caused it, but the behaviour did not change after removing keymaps. On my computer the scripts are executed in this order: After the key mappings are loaded, lo is brought up, after which the random number generator is initialized and runlevel 3 is entered...
You might want to play with /sbin/splash-functions.sh a little to make detecting the source of this problem easier. You could try adding something like 'echo "STARTUP SERVICE ERROR: ${srv}"' on line 228 of that file (the body of splash_svc() function, just before '/sbin/splash "verbose"'. The part of the code after the modifications would read as follows: if [ "${err}" -ne 0 -a "${SPLASH_VERBOSE_ON_ERRORS}" = "yes" -a "${srv}" != "consolefont" ]; then echo "STARTUP SERVICE ERROR: ${srv}" /sbin/splash "verbose" return 1 fi After that, you would get a nice big error message with the name of the service that is causing the switch to verbose. If you decide to try it, please let me know which service is it.
Thanks, Michal! It was the script "serial" which produced the error. I think it did not print an error msg by itself because I neither have serial support built in my kernel nor loaded as a module. Do you think this is something which should be fixed in the "serial" script? Best regards Christian
It could very well be, but you might want to try upgrading baselayout to the latest "unstable" version. If the problem remains, we'll look into it. I have the latest available baselayout installed and I can't reproduce the problem with /etc/init.d/serial returning a non-zero exit code (and I don't have any serial devices or serial.o module either).
I have got the same problem as Christian. Removing "serial" from boot (rc-update del serial) resolved the issue. There seems to be a problem in the serial script! I am running the latest unstable unmasked baselayout-1.11.10-r7.
