| Summary: | bootsplash minor correction in local.start | ||
|---|---|---|---|
| Product: | Gentoo Linux | Reporter: | Arun Bhanu (RETIRED) <arunbhanu> |
| Component: | Current packages | Assignee: | Troy Dack <troy> |
| Status: | RESOLVED FIXED | ||
| Severity: | minor | ||
| Priority: | High | ||
| Version: | unspecified | ||
| Hardware: | x86 | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Package list: | Runtime testing required: | --- | |
| Attachments: |
local.start.patch
bootsplash-0.4.ebuild.diff /etc/init.d/bootsplash |
||
|
Description
Arun Bhanu (RETIRED)
2003-03-18 11:49:41 UTC
Created attachment 9565 [details, diff]
local.start.patch
this one line patch should fix the problem
actually, I am going to release a -r1 for bootsplash with its own initscripts. it should not use local Created attachment 9582 [details, diff]
bootsplash-0.4.ebuild.diff
Patch against bootsplash-0.4.ebuild to include rc script for bootsplash and to
install bootsplash into /usr/share/${PN} instead of /usr/share/${P}
Created attachment 9583 [details]
/etc/init.d/bootsplash
/etc/init.d rc script for bootsplash - to be plased in ${FILESDIR}.
Please test this and the preceeding patch.
Note: I have not tested these as they were produced under Cygwin with the aid
of my very unreliable memory and the very reliable Gentoo docs :)
For information: The rc script will leave the VT's looking a little "untidy", especially if you are using something like linux_logo to create a fancy /etc/issue. This can be fixed by doing a "killall agetty" (probably a SIGHUP to each agetty will do) after changing the frame buffer image. Since the agetty's are all listed in /etc/inittab with respawn then they simply restart and re-display /etc/issue, taking into account the changed terminal size (columns & rows) From memory the killall will respond with something about no processes being killed, then the /etc/issue contents is displayed. OK,
New and improved 0.4-r1 ebuild commited.
Now has /etc/init.d/bootsplash rc script, no more hacking
/etc/conf.d/local.{start,stop}
Arun, in regard to your initial comment:
There are two lots of config files and images. One is for the actual
startup/shutdown of Gentoo, while all the Starting ..... [ OK ] messages are
displayed. (the bootsplash.{cfg,jpg} symlinks)
The second set of images are for the consoles when they are in normal use.
(the consolesplash.{cfg,jpg} symlinks)
The initrd is created using the bootsplash.* symlinks.
Please test and let me know how it goes.
I like your idea of having two symbolic links consolesplash & bootsplash. okay, i've tested the bootsplash-0.4-r1 with gentoo-sources-2.4.20-r2. Everything is working fine except for a very trivial problem. Couple of things I would like to clarify: I had to manually delete the entries in local.start & local.stop. Is there a way we could provide ._cfg in diff format so that users upgrading from 0.4 to 0.4-r1 can do a etc-update and doesn't have to manually update those files. Just an idea. I don't know whether it is difficult to achieve. There is minor problem though. If you use two cfg files - one for bootsplash and one for consolesplash with different values for tw then I encounter the follwing behaviour. bootsplash.cfg contains tw=780 and consolesplash contains tw=1008. When I reboot, I get the messages like * Setting Console frame buffer images [ok] * Setting local [ok] All other messages after that align properly. Notice that [ok] is in the next line. But if both cfg files contain the same tw then when you reboot I get all the messages get aligned properly. * Setting Console frame buffer images [ok] * Setting local [ok] other messages... Can anyone else reproduce this or is it just me? Arun,
bootsplash has now made it to 0.5. A -r1 will be out shortly to address a couple of minor problems that 0.5 has.
I'm not sure if it is possible to provide a ._cfg diff, I'll see what I can do.
Regarding the formatting of the [ OK ] messages, I don't think there is a lot too be done there. The rc scripts caclulate the terminal width before the width of the terminal is changed by changing the console image. This is why the bootsplash message is formatted incorrectly and all the other messages come of properly.
|