2.6.7-love8 aka back from the hols

Message

boroshan · Post by **boroshan** » Mon Jul 12, 2004 9:08 am

Frodg wrote:
PickledOnion wrote:
Check the link, that is to love7 not love8
Yes - noticed that so I went back to the original link (page 1) and got that one.... I assume it is OK - I have not seen others with this problem.

I get this semi-regularly downloading love. It happens when the fetch bombs out early for whatever reason. Given that some of the servers hostling love patches were never intended to support this sort of bandwidth, it's not surprising that it's happened. Hopefully it'll be less of a problem with sourceforge hosting from now on.

The problem seems to be complicated by the digest files, which appear to tell portage what length of file to expect. So, if you generated digests on a short file, the next time portage fetches it, it fetches it short. In any event, clobbering the digest files in

Code: Select all

/usr/local/portage/sys-kernel/love-sources/files/*

seems to fix it for me

Pink · Post by **Pink** » Mon Jul 12, 2004 9:37 am

boroshan: You may indeed be right I have never had that problem personally though.

The main reason I know that link is to love7 and not love8 is that it is my server (pickledonion.net) and it is the only love I am hosting at the moment

boroshan · Post by **boroshan** » Mon Jul 12, 2004 10:26 am

I think it may happen when I'm at my gf's place. She's on dialup and Tiscali kicks us off every couple of hours.

It's no big deal, just a minor irritation. The best bet is often to manually resume the fetch with wget -c

Frodg · Post by **Frodg** » Mon Jul 12, 2004 11:41 pm

boroshan wrote:
Frodg wrote:
PickledOnion wrote:
Check the link, that is to love7 not love8
Yes - noticed that so I went back to the original link (page 1) and got that one.... I assume it is OK - I have not seen others with this problem.
I get this semi-regularly downloading love. It happens when the fetch bombs out early for whatever reason. Given that some of the servers hostling love patches were never intended to support this sort of bandwidth, it's not surprising that it's happened. Hopefully it'll be less of a problem with sourceforge hosting from now on.

The problem seems to be complicated by the digest files, which appear to tell portage what length of file to expect. So, if you generated digests on a short file, the next time portage fetches it, it fetches it short. In any event, clobbering the digest files in
Code: Select all
/usr/local/portage/sys-kernel/love-sources/files/*
seems to fix it for me

Worked perfectly

Brilliant

Frodg · Post by **Frodg** » Mon Jul 12, 2004 11:57 pm

I now have love 8 working but I still have a problem that I have had with loves 6 & 7 ( I was hoping a fresh love 8 would fix it) :

Slow internet

I know it has been discussed in the forums for those love releases but the fixed /patches (evernmind, etc) did not work for me.

I get a fast immediate download - around 100 to 200 K/s or higher depending on the site - and then 10 seconds in it slows down to around 800 B/s (takes me back to the 9.6k modem days

)

I have cable and if I reboot in 2.6.7 love 5 - my current default - I get back the speed

Any ideas - any logs I could post ?

MrApples · Post by **MrApples** » Tue Jul 13, 2004 12:02 am

what driver are you using? ive heard a lot of this dealing with tulip

Frodg · Post by **Frodg** » Tue Jul 13, 2004 12:40 am

No, Its the 8129too - realtek....

nmcsween · Post by **nmcsween** » Tue Jul 13, 2004 1:01 am

This happens with the forcedeth driver for any nforce boards as well. Might be a bug in the kernel since this has been happening to me since 2.6.5

nuzzy · Post by **nuzzy** » Tue Jul 13, 2004 1:20 am

for those who got the ebuild "html-ized"....just "view source" in your browser options and copy/paste into an editor and make sure it's named correctly

phlashback · Post by **phlashback** » Tue Jul 13, 2004 1:24 am

my bad about posting the wrong link

I copied it off the the info from #love-sources, and it was listed as love 8 there (I think).. next time I will verify. Busy building a new server, and paying close attention to details there

Taken from the sorceforge site

akimoto · Post by **akimoto** » Tue Jul 13, 2004 5:20 am

Lovin' it

m0sia · Post by **m0sia** » Tue Jul 13, 2004 10:41 am

i have a lot of problems:
1) botsplash don't work. vesafb works corrrect, but no bootsplash. btw i don't care about it, but i don't like it.

2) usb mouse & usb flash disk don't work. i think, that all usb subsystem don't work. all necessary modules inserted with hotplug.

patch mensioned above don't work for me, because it is already in kernel, i've downloaded.

3) cupsd freezes at startup. it is very annoying.
i tried patch, found on the forum, but it doesn't help =(
btw i tried acpi disabled but it still don't work.

4) systrem hangs at reboot or poweroff. even if i disable acpi.

now i downgraded to 2.6.7-love5, which i used before.

ps i wonder if it is possibly to include swsup2 in love sources? i'd like to test it.

Dinini · Post by **Dinini** » Tue Jul 13, 2004 12:45 pm

Nick Piggin wrote:http://www.kerneltrap.org/~npiggin/2.6.7-np8/

A few important fixes to the memory management stuff, and fixed
a regression in scheduler interactivity.

If applying to Love8
get:
http://www.kerneltrap.org/~npiggin/2.6.7-mm6-np5.gz
http://www.kerneltrap.org/~npiggin/2.6. ... mm7-np8.gz

Code: Select all

cd /usr/src/linux-2.6.7-love8/
gzcat /path/to/2.6.7-mm6-np5.gz | patch -Rp1
gzcat /path/to/2.6.7-mm7-np8.gz | patch -p1

The one reject on reversing 2.6.7-mm6-np5.gz can be ignored.
The 2 rejects on applying 2.6.7-mm7-np8.gz need to fixed (the vmscan.c file does have a 1 line change from what was in love8).

Assuming you trust me to have fixed the rejects properly

get:
http://ljones.com/2.6.7-love8-np8-rej-fix.patch

Code: Select all

cd /usr/src/linux-2.6.7-love8/
cat /path/to/2.6.7-love8-np8-rej-fix.patch | patch -p0

The patch worked ok here, but the power outage delayed posting the patch. I love rural power companies... no I don't.

MrApples · Post by **MrApples** » Tue Jul 13, 2004 3:55 pm

thanks dinini

/me starts patching

squeegy · Post by **squeegy** » Tue Jul 13, 2004 4:33 pm

m0sia wrote:i have a lot of problems:
1) botsplash don't work. vesafb works corrrect, but no bootsplash. btw i don't care about it, but i don't like it.

2) usb mouse & usb flash disk don't work. i think, that all usb subsystem don't work. all necessary modules inserted with hotplug.

patch mensioned above don't work for me, because it is already in kernel, i've downloaded.

3) cupsd freezes at startup. it is very annoying.
i tried patch, found on the forum, but it doesn't help =(
btw i tried acpi disabled but it still don't work.

4) systrem hangs at reboot or poweroff. even if i disable acpi.

now i downgraded to 2.6.7-love5, which i used before.

ps i wonder if it is possibly to include swsup2 in love sources? i'd like to test it.

did you try reversing bk-acpi.patch and bk-usb.patch, that fixes all the problems i had which were very similar to yours. I posted how to do this in one of the love threads, not sure if it was this one or not.

squeegy · Post by **squeegy** » Tue Jul 13, 2004 5:06 pm

Okay so anyone experiencing problems with Java applications just dying, I found a fix that works for at least one person. Here are the steps to apply the reverse patch.

Code: Select all

cd /usr/src
wget http://linux.bkbits.net:8080/linux-2.6/gnupatch@40d35213GZJkY_U5Nc0ugteABgtjHA
cd linux (assuming love8 is symlinked to it)
patch -p1 -R < ../gnupatch@40d35213GZJkY_U5Nc0ugteABgtjHA

that should be at least a decent temporary fix for the problem.

evermind · Post by **evermind** » Tue Jul 13, 2004 5:37 pm

oberyno wrote:Any chance of bringing back badram? I've got 4k of bad ram which of course screws up everything.
I tried patching it myself, but had no idea what I was doing trying to fix the reject.

just type

Code: Select all

cat BadRAM-2.6.5.2.patch |patch -p1 --dry-run -F3

and it will patch without rejects

EDIT:
--dry-run is just to test if the patch would work
remove it to really patch

aent · Post by **aent** » Tue Jul 13, 2004 5:49 pm

Can anyone get swsusp2 patches to apply to this?

MighMoS · Post by **MighMoS** » Tue Jul 13, 2004 6:16 pm

I'm getting

Code: Select all

make[1]: `arch/i386/kernel/asm-offsets.s' is up to date.
  CHK     include/linux/compile.h
  CC      kernel/sched.o
kernel/sched.c: In function `rebalance_tick':
kernel/sched.c:1888: error: 'rq' redeclared as different kind of symbol
kernel/sched.c:1885: error: previous definition of 'rq' was here
kernel/sched.c:1894: error: `p' undeclared (first use in this function)
kernel/sched.c:1894: error: (Each undeclared identifier is reported only once
kernel/sched.c:1894: error: for each function it appears in.)
kernel/sched.c:1902: error: `EXIT_WEIGHT' undeclared (first use in this function)
make[1]: *** [kernel/sched.o] Error 1
make: *** [kernel] Error 2

Any ideas?

dr_strange · Post by **dr_strange** » Tue Jul 13, 2004 6:29 pm

Frodg wrote:No, Its the 8129too - realtek....

Experiencing the same effect, unfortunately.

MrApples · Post by **MrApples** » Tue Jul 13, 2004 6:34 pm

dr_strange wrote:
Frodg wrote:No, Its the 8129too - realtek....
Experiencing the same effect, unfortunately.

i cant believe this problem hasnt been solved yet, its all over the forums

Frodg · Post by **Frodg** » Tue Jul 13, 2004 9:58 pm

MrApples wrote:
dr_strange wrote:
Frodg wrote:No, Its the 8129too - realtek....
Experiencing the same effect, unfortunately.
i cant believe this problem hasnt been solved yet, its all over the forums

Yeh,

Thats what I thpought but none of the patches /suggestions work..

m0sia · Post by **m0sia** » Wed Jul 14, 2004 3:57 am

[quote="squeegy"]
did you try reversing bk-acpi.patch and bk-usb.patch, that fixes all the problems i had which were very similar to yours. I posted how to do this in one of the love threads, not sure if it was this one or not.[/quote]
of cource i tried reversing bk-acpi.patch and bk-usb.patch, as you mentioned in this thread, but problem still exist time to time: some times it boots perfectly, sometimes with this problems. But when it boots it "ROX LIKE NINJA" =)

oberyno · Post by **oberyno** » Wed Jul 14, 2004 4:38 am

evermind wrote:just type
Code:
cat BadRAM-2.6.5.2.patch |patch -p1 --dry-run -F3

and it will patch without rejects

Thanks. That fuzz factor was what I needed.

_Adik_ · Post by **_Adik_** » Wed Jul 14, 2004 5:11 am

2.6.8-rc1-mm1 is out!

so we're waiting for the new love...