2.6.2_rc2-love4 aka "Jimmy Has A Cheeto"

Message

petrjanda · Post by **petrjanda** » Sat Jan 31, 2004 9:18 am

can anyone tell me if its worth upgrading to love4 from love1?

J0rus · Post by **J0rus** » Sat Jan 31, 2004 11:24 am

Evil Dark Archon wrote:finally found the real culprit, the new acpi patch in mm2 is what was causing the problem, it kept giving errors about i8xxx(don't remember the numbers i know it has something to do with the ps/2 mouse port) and disabling interrupt 12 and also the siimage lost interrupt problem. Now that all that is sorted out (by disabling acpi) its running good, and as an added bonus its doing so faster than -love3.

I too was having issues with the new ACPI patch; lost interrupt errors with my promise controller and no network.
Tracked it down to the new MMConfig option under PCI Access mode.
I selected 'Direct' as opposed to 'Any', effectively bypassing the MMConfig detection attempt. Came up clean after a recompile, no further issues as of yet.

BTW -- Nick's VM swappiness makes a HUGE difference!!

~J0rus

OneOfOne · Post by **OneOfOne** » Sat Jan 31, 2004 12:08 pm

1 major problem with all rc2-love* and 1 with love4 (-mm2).
1. my harddisk is AWFULLY slow, 16-23mb instead of 47mb... (kinda tracked it down to intel mobo but can't tell for sure)
2. on -mm2 i have to pass acpi=off to the kernel or my modem will oops (speedtouch adsl 330 usb).
back to my good ol' 2.6.2-rc1-love1

peace

grzewho · Post by **grzewho** » Sat Jan 31, 2004 12:16 pm

the vesafb driver (or bootsplash, dunno) gives lots of garbage output to syslog

any fixes ?

DarwinianCoeus · Post by **DarwinianCoeus** » Sat Jan 31, 2004 1:18 pm

neenee · Post by **neenee** » Sat Jan 31, 2004 1:28 pm

is out

luqas · Post by **luqas** » Sat Jan 31, 2004 1:46 pm

yes!!! play time again soon.... Considering I just jumped into the love sources not to long ago, I am itching for a new release already

.

Yen · Post by **Yen** » Sat Jan 31, 2004 4:32 pm

How do I install this patched kernel?

thundersteele · Post by **thundersteele** » Sat Jan 31, 2004 6:58 pm

@steel300

I would like to see a line in your first post and perhaps in notes.txt where it says which kernel source your patch applies to.
When it started lovechilds patches were based on latest mm-patches. I don't know if this changed someday. So now I don't know if your patch applies to latest stable 2.6 kernel, latest prepatch or latest mm-sources. Does rc2 tell me it is for 2.6.2-rc2?

Thanx for your help, and for this great work.

nepenthe · Post by **nepenthe** » Sat Jan 31, 2004 7:54 pm

thundersteele,

from the notes.txt

Patch Name: patch-2.6.2-rc2
Description: The latest prepatch for the 2.6 tree
Testing: Patch applies cleanly and compiles cleanly

Patch Name: 2.6.2-rc2-mm1
Description: The latest from Andrew Morton
Testing: Patch applies cleanly and compiles cleanly

So you aplly it to 2.6.1 or you could just use the provided ebuild =)

pdptrojan · Post by **pdptrojan** » Sat Jan 31, 2004 7:56 pm

With this kernel, my hard driver thrashes like crazy and i have very poor responsivness when at 100% cpu. This didn't happen in the 2.6.2_rc1 love sources. Like some other people, an hdparm -tT shows terrible speeds in comparison before (and yes, all the dma etc. is working). Anybody know why this is or what I can do to fix it?

ed0n · Post by **ed0n** » Sat Jan 31, 2004 8:54 pm

pdptrojan wrote:With this kernel, my hard driver thrashes like crazy and i have very poor responsivness when at 100% cpu. This didn't happen in the 2.6.2_rc1 love sources. Like some other people, an hdparm -tT shows terrible speeds in comparison before (and yes, all the dma etc. is working). Anybody know why this is or what I can do to fix it?

me too and OneOfOne too . I guess that we are using intel motherboards.
That's why I also got back to love1.

MG-Cloud · Post by **MG-Cloud** » Sat Jan 31, 2004 8:59 pm

Hi,

I've had some problems with some of Nicks schedulers in the past - no big deal but I was wondering if they are present in the current version or if there's a workaround.

I currently need to run starcraft nice'd:

nice -n20 wine starcraft.exe

This is so that it won't lag on 2 player games and also so that it'll be much more responsive. Unfortunately, with some of the earlier versions of love sources that included nicks scheduler (not all of them), running it nice'd had a detrimental effect to performance. Running it just like wine starcraft.exe seemed to be okay for 3 player games, but I was unable to test it for larger games before updating to the newest love-sources kernel without this scheudler

d99ma · Post by **d99ma** » Sat Jan 31, 2004 9:50 pm

ed0n wrote:
pdptrojan wrote:With this kernel, my hard driver thrashes like crazy and i have very poor responsivness when at 100% cpu. This didn't happen in the 2.6.2_rc1 love sources. Like some other people, an hdparm -tT shows terrible speeds in comparison before (and yes, all the dma etc. is working). Anybody know why this is or what I can do to fix it?
me too and OneOfOne too . I guess that we are using intel motherboards.
That's why I also got back to love1.

I also have this problem, anyone tried vanilla 2.6.2-rc2 or mm to se where the problem was added?

Master_Of_Disaster · Post by **Master_Of_Disaster** » Sat Jan 31, 2004 9:55 pm

@steel
Could you please provide a broken-out directory of your patches? I'd like to experiment a bit with kernel patches, and I'd like to use some of yours too.
Thanks in advance!

Yen · Post by **Yen** » Sat Jan 31, 2004 10:09 pm

I mean, where do i need to move the ebuild file to?
What command do i need to type?

Master_Of_Disaster · Post by **Master_Of_Disaster** » Sat Jan 31, 2004 10:14 pm

move it to $PORTDIR_OVERLAY/sys-kernel/love-sources, then issue as root:

$ ebuild /path/to/ebuild digest
$ emerge love-sources

nsahoo · Post by **nsahoo** » Sun Feb 01, 2004 12:56 am

nsahoo wrote:my mouse is still getting stuck. it happens every 5-10 mins. when i unplug and plug the mouse again, it starts working but for next 10 mins or so. I am using MS intellimouse (optical, USB, 5 button). Any help?

Evil Dark Archon wrote:are you sure that you have eliminated all other possible problems other than the kernel, it may be a problem with the mouse itself, does it work if you boot into a previous version, could you post your .config file.

I switched to 2.6.0-test11 and the problem was gone. Now I am in 2.6.1-gentoo-r1 and the problem hasn't happened in last 15-20 mins. May be one of these patches did it.

ett_gramse_nap · Post by **ett_gramse_nap** » Sun Feb 01, 2004 9:50 am

I'm compiling now, but I did't notice any options about that "nick swappiness"-thing Steel mentioned. Is it enabled anyway? And what is it? A search here on the forums only returns the love-threads...

silverter · Post by **silverter** » Sun Feb 01, 2004 10:05 am

d99ma wrote:
ed0n wrote:
pdptrojan wrote:With this kernel, my hard driver thrashes like crazy and i have very poor responsivness when at 100% cpu. This didn't happen in the 2.6.2_rc1 love sources. Like some other people, an hdparm -tT shows terrible speeds in comparison before (and yes, all the dma etc. is working). Anybody know why this is or what I can do to fix it?
me too and OneOfOne too . I guess that we are using intel motherboards.
That's why I also got back to love1.
I also have this problem, anyone tried vanilla 2.6.2-rc2 or mm to se where the problem was added?

I noticed the same problem in 2.6.2-rc2-mm1, will be trying vanilla and see if the problem comes from there or if it was introduced in mm.

ett_gramse_nap · Post by **ett_gramse_nap** » Sun Feb 01, 2004 10:50 am

Works like a charm over here! Had som trouble steeming from disabling systemvpl (or something like that) with X but that isn't a love-issue...

Still curious about what I wrote in my earlier post thou... Can someone please explain or posts some links regarding nick's swappiness stuff!

ProtectionFault · Post by **ProtectionFault** » Sun Feb 01, 2004 11:34 am

2.6.2_rc2-love4 rocks!!!!!!!!!!!
Love sources rule...

Broken out directory would be great!

thx in advance

greetz

ProtectionFault

_Adik_ · Post by **_Adik_** » Sun Feb 01, 2004 4:16 pm

houston, we have a little problem here:

Code: Select all

  CHK     include/linux/compile.h
  LD      init/built-in.o
  LD      arch/i386/kernel/built-in.o
  CC [M]  drivers/net/r8169.o
drivers/net/r8169.c:420: `RxUnderrun' undeclared here (not in a function)
drivers/net/r8169.c: In function `rtl8169_print_mac_version':
drivers/net/r8169.c:509: warning: implicit declaration of function `dprintk'
drivers/net/r8169.c: In function `rtl8169_phy_timer':
drivers/net/r8169.c:677: `RTL8169_PHY_TIMEOUT' undeclared (first use in this function)
drivers/net/r8169.c:677: (Each undeclared identifier is reported only once
drivers/net/r8169.c:677: for each function it appears in.)
drivers/net/r8169.c: In function `rtl8169_request_timer':
drivers/net/r8169.c:706: `RTL8169_PHY_TIMEOUT' undeclared (first use in this function)
drivers/net/r8169.c: In function `rtl8169_init_one':
drivers/net/r8169.c:880: `RTL8169_TX_TIMEOUT' undeclared (first use in this function)
drivers/net/r8169.c: In function `rtl8169_open':
drivers/net/r8169.c:1104: `R8169_TX_RING_BYTES' undeclared (first use in this function)
drivers/net/r8169.c:1109: `R8169_RX_RING_BYTES' undeclared (first use in this function)
drivers/net/r8169.c: In function `rtl8169_hw_start':
drivers/net/r8169.c:1155: `EarlyTxThres' undeclared (first use in this function)
drivers/net/r8169.c:1155: `EarlyTxThld' undeclared (first use in this function)
drivers/net/r8169.c: In function `rtl8169_try_rx_copy':
drivers/net/r8169.c:1466: `rx_copybreak' undeclared (first use in this function)
drivers/net/r8169.c: In function `rtl8169_interrupt':
drivers/net/r8169.c:1584: `priv' undeclared (first use in this function)
drivers/net/r8169.c: In function `rtl8169_close':
drivers/net/r8169.c:1622: structure has no member named `phy_timer_t'
drivers/net/r8169.c:1646: structure has no member named `TxDescArrays'
drivers/net/r8169.c:1647: structure has no member named `RxDescArrays'
drivers/net/r8169.c:1648: structure has no member named `TxDescArrays'
drivers/net/r8169.c:1649: structure has no member named `RxDescArrays'
drivers/net/r8169.c:1656: warning: implicit declaration of function `RTL8169_FREE_RXSKB'
drivers/net/r8169.c: In function `rtl8169_set_rx_mode':
drivers/net/r8169.c:1721: warning: passing arg 2 of `set_bit' from incompatible pointer type
drivers/net/r8169.c: At top level:
drivers/net/r8169.c:1046: warning: `rtl8169_suspend' defined but not used
drivers/net/r8169.c:1072: warning: `rtl8169_resume' defined but not used
make[2]: *** [drivers/net/r8169.o] Error 1
make[1]: *** [drivers/net] Error 2
make: *** [drivers] Error 2

and it stops...

any ideas??

steveb · Post by **steveb** » Sun Feb 01, 2004 5:16 pm

i tryed the love4 sources on 2 servers and both of them have problems with sunrpc and nfsd modules.
the love3 sources did not have that problem!
does any1 else have the same problem?

cheers

SteveB

d99ma · Post by **d99ma** » Sun Feb 01, 2004 9:34 pm

silverter wrote:
d99ma wrote:
ed0n wrote:
pdptrojan wrote:With this kernel, my hard driver thrashes like crazy and i have very poor responsivness when at 100% cpu. This didn't happen in the 2.6.2_rc1 love sources. Like some other people, an hdparm -tT shows terrible speeds in comparison before (and yes, all the dma etc. is working). Anybody know why this is or what I can do to fix it?
me too and OneOfOne too . I guess that we are using intel motherboards.
That's why I also got back to love1.
I also have this problem, anyone tried vanilla 2.6.2-rc2 or mm to se where the problem was added?
I noticed the same problem in 2.6.2-rc2-mm1, will be trying vanilla and see if the problem comes from there or if it was introduced in mm.

I'm going to try the gentoo-dev-sources and se how they work. Keep us posted on vanilla-performance. Seems unlikley that such a noticable bug would be in the main tree though.

2.6.2_rc2-love4 aka "Jimmy Has A Cheeto"

oops...