Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
Love-sources - get them before they go obsolete
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next  
Reply to topic    Gentoo Forums Forum Index Off the Wall
View previous topic :: View next topic  
Author Message
bssteph
l33t
l33t


Joined: 26 Feb 2003
Posts: 651
Location: Wisconsin

PostPosted: Sat Sep 13, 2003 2:29 pm    Post subject: Reply with quote

Lovechild wrote:
Love gets whooped in the IO tests because -mm1 and vanilla uses the AS scheduler which is blazing fast for IO - where as -love uses CFQ per default, which aims for complete fairness.

If you rerun the test with elevator=as appended to the -love kernel you should see improvements in that area.


Yeah, but the odd thing is that it's only -love1... -love2 was still able to pick up on the load well enough to become fair and still squeak out mm1. So that was a bit odd. Either way, love's scoring well. :)

Two machines (the desktop and the contest laptop) both compiling -love3, I'll append contest results if I get to finish them this morning. Maybe I'll toss in a elevator=cfq on mm1 too, just to get a better comparison.
Back to top
View user's profile Send private message
Wedge_
Advocate
Advocate


Joined: 08 Aug 2002
Posts: 3614
Location: Scotland

PostPosted: Sat Sep 13, 2003 2:50 pm    Post subject: Reply with quote

Running -love3 now, everything seems fine. The latest reiser4progs won't compile for me though, I don't know if something got screwed up when I changed the ebuilds..
Code:
gcc -DHAVE_CONFIG_H -I. -I. -I../../.. -I../../../include -DENABLE_STAND_ALONE -DENABLE_SYMLINKS -DENABLE_R5_HASH -DENABLE_DEG_HASH -falign-jumps=1 -falign-loops=1 -falign-functions=1 -falign-labels=1 -fomit-frame-pointer -fno-inline -fstrength-reduce -DENABLE_LARGE_KEYS -Os -W -Wall -Wuninitialized -Wno-unused-parameter -Wredundant-decls -MT libdir40_alone_la-dir40.lo -MD -MP -MF .deps/libdir40_alone_la-dir40.Tpo -c dir40.c  -fPIC -DPIC -o .libs/libdir40_alone_la-dir40.lo
dir40.c: In function `dir40_update':
dir40.c:214: error: `units' undeclared (first use in this function)
dir40.c:214: error: (Each undeclared identifier is reported only once
dir40.c:214: error: for each function it appears in.)
dir40.c: At top level:
dir40.c:329: warning: `dir40_compent' defined but not used
make[4]: *** [libdir40_alone_la-dir40.lo] Error 1
make[4]: Leaving directory `/var/tmp/portage/reiser4progs-0.4.14/work/reiser4progs-0.4.14/plugin/object/dir40'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/var/tmp/portage/reiser4progs-0.4.14/work/reiser4progs-0.4.14/plugin/object'
make[2]: *** [all-recursive] Error 1
make[2]: Leaving directory `/var/tmp/portage/reiser4progs-0.4.14/work/reiser4progs-0.4.14/plugin'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/var/tmp/portage/reiser4progs-0.4.14/work/reiser4progs-0.4.14'
make: *** [all] Error 2

!!! ERROR: sys-apps/reiser4progs-0.4.14 failed.
!!! Function src_compile, Line 43, Exitcode 2
!!! Failed to compile


Edit: fixed by change --disable-collisions to --enable-collisions in the ebuild, but I don't know what that does :)
_________________
Per Ardua Ad Astra
The Earth is the cradle of the mind, but we cannot live forever in a cradle - Konstantin E. Tsiolkovsky
Gentoo Radeon FAQ
Back to top
View user's profile Send private message
Safrax
Guru
Guru


Joined: 23 Apr 2002
Posts: 422

PostPosted: Sat Sep 13, 2003 4:01 pm    Post subject: Reply with quote

How do I go about compiling this with icc? I tried changing all instances of gcc to icc and g++ to icpc in the Makefile. It fails on make oldconfig with a bunch of error stuff.
Back to top
View user's profile Send private message
bssteph
l33t
l33t


Joined: 26 Feb 2003
Posts: 651
Location: Wisconsin

PostPosted: Sat Sep 13, 2003 4:30 pm    Post subject: Reply with quote

-love3 results are in.
http://bssteph.irtonline.org/linux/testing/2.6.0-test5_vanilla_mm1-O20.1_love.txt

I wasn't expecting much of a change from love2 since Nick v16 wasn't in, but this surprised me in one of those "... oh." kind of ways.
Tests either were roughly equivalent to -love2 or dropped slightly, with a significant drop in dbench... so significant that -mm1 does better and it makes me question my results.

I kept the same config so it isn't OOM.. compiled with gcc 3.3.1
Back to top
View user's profile Send private message
sindre
Guru
Guru


Joined: 01 Nov 2002
Posts: 315
Location: Norway

PostPosted: Sat Sep 13, 2003 5:03 pm    Post subject: Reply with quote

Any of the results that's more significant on the desktop than others?
Adding all together you get this (assuming I understood the results):
Code:
1. 2.6.0-test5-love2  16.25
2. 2.6.0-test5-mm1    17.96
3. 2.6.0-test5-love1  18.31
4. 2.6.0-test5-love3  19.36
5. 2.6.0-test5        21.68
Back to top
View user's profile Send private message
bssteph
l33t
l33t


Joined: 26 Feb 2003
Posts: 651
Location: Wisconsin

PostPosted: Sat Sep 13, 2003 5:32 pm    Post subject: Reply with quote

For an overall "big picture" you can add the ratios together for a generic comparison, but it of course doesn't tell the entire story.

Con writes (from the contest page)
Quote:
The lower the time (and ratio) the better, and the higher the cpu percentage the
better. The absolute number is not important, but the differences are.
When the difference between noload and the loads is small it shows that
on that kernel, the system is able to respond to requests for normal
tasks (ie responsiveness).


It's hard to say if any of the tests are more significant on desktop than others... they represent things you could easily be doing on a desktop, just done over and over as part of the test.
Back to top
View user's profile Send private message
Lovechild
Advocate
Advocate


Joined: 17 May 2002
Posts: 2858
Location: Århus, Denmark

PostPosted: Sat Sep 13, 2003 8:33 pm    Post subject: Reply with quote

I am surprised at the difference between love2 and love3 since nothing I included should affect contest.

Especially dbench should be equally great, as CFQ hasn't changed at all.

Is contest prone to statistical errors ??
Back to top
View user's profile Send private message
dedeaux
Apprentice
Apprentice


Joined: 19 Jun 2002
Posts: 182
Location: Songtan, South Korea

PostPosted: Sun Sep 14, 2003 5:26 am    Post subject: should one try this? Reply with quote

I was wondering, is it advisable to try this out? I use my laptop with linux and have Gentoo running nicely, but love to explore new things. I just recovered well from the move to the BMG Gnome 2.4 and am tempted by this thread.

Should I take the plunge? I know there are several howtos in the forum, but I wanted to know who uses 2.6 full-time.

Also, is there wireless support in this kernel?
Back to top
View user's profile Send private message
Safrax
Guru
Guru


Joined: 23 Apr 2002
Posts: 422

PostPosted: Sun Sep 14, 2003 7:51 am    Post subject: Reply with quote

I've used 2.6 full time since 2.5.63 something.. It has everything 2.4 has and more for the most part.


BTW, I'm still trying to get this to compile with icc. I haven't been able to find out anything about the patch lovechild applied so I haven't been able to get very far.. heh
Back to top
View user's profile Send private message
Lovechild
Advocate
Advocate


Joined: 17 May 2002
Posts: 2858
Location: Århus, Denmark

PostPosted: Sun Sep 14, 2003 8:12 am    Post subject: Reply with quote

the patch basically just "bypasses" the inlined asm which icc can't understand - the Intel engineers say they can compile the kernel with their compiler now (Linus used frases such as broken about icc btw.), there's a better patch coming I hear.

How you compile it with ICC I dunno, search lkml or google :)
Back to top
View user's profile Send private message
bssteph
l33t
l33t


Joined: 26 Feb 2003
Posts: 651
Location: Wisconsin

PostPosted: Sun Sep 14, 2003 8:19 am    Post subject: Reply with quote

I don't know what the hell's going on with contest. Before I left today I fired off a run with -mm1 while using cfq. In some areas its results looked the same as mm1 using AS, others were sporatically different. Loads similar to mm1 when I expected something closer to love due to using the different elevator. A huge difference between io_load and io_other results.
I'm not a guru of contest but I'm baffled.

Tomorrow morning I'm going to try a bit of cleanup on the partition and then a fresh untar of 2.4.19 and i'll rerun a couple tests just to make sure something hasn't gone nutty. And we'll see then what this all means. I think I may make special versions of my kernels w/framebuffer and acpi/apm off just to make extra sure.

Subjectively, though, love3 feels nice Lovechild. I didn't get much of a chance to do stress testing (or any really), but as you said not much of anything changed that would warrant them. Eagerly awaiting v16 and the resulting love set. :)
Back to top
View user's profile Send private message
Safrax
Guru
Guru


Joined: 23 Apr 2002
Posts: 422

PostPosted: Sun Sep 14, 2003 8:25 am    Post subject: Reply with quote

Ahh well I suppose I'll just give up for now. I do hope intel hurrys up and whips out a gcc compatible ICC. My p4 laptop will thank me as well as my athlon desktop if ATi ever gets even remotely close to nvidias support.
Back to top
View user's profile Send private message
Lovechild
Advocate
Advocate


Joined: 17 May 2002
Posts: 2858
Location: Århus, Denmark

PostPosted: Sun Sep 14, 2003 10:07 am    Post subject: Reply with quote

Remember -mm has some extensive AIO work, and other nice stuff, I still think -love "feels" better though... but my opinion is tainted I guess.

I have a few tricks up my sleeve for -love4 - but most likely I'll back out a lot of the useless changes so we again have a minimal kernel.
Back to top
View user's profile Send private message
Lovechild
Advocate
Advocate


Joined: 17 May 2002
Posts: 2858
Location: Århus, Denmark

PostPosted: Sun Sep 14, 2003 10:24 am    Post subject: Re: should one try this? Reply with quote

dedeaux wrote:
I was wondering, is it advisable to try this out? I use my laptop with linux and have Gentoo running nicely, but love to explore new things. I just recovered well from the move to the BMG Gnome 2.4 and am tempted by this thread.

Should I take the plunge? I know there are several howtos in the forum, but I wanted to know who uses 2.6 full-time.

Also, is there wireless support in this kernel?


Go for it !!!
Back to top
View user's profile Send private message
neenee
Veteran
Veteran


Joined: 20 Jul 2003
Posts: 1786

PostPosted: Sun Sep 14, 2003 10:31 am    Post subject: Reply with quote

i can vouch for this patch as well. i've been following
it for a bit now, and have used it for work and play.
_________________
proud to be a scout and a chronic penguin hugger
Legion of Lore - site
Back to top
View user's profile Send private message
zypher
Retired Dev
Retired Dev


Joined: 10 Jun 2002
Posts: 416
Location: Cologne, ger.

PostPosted: Sun Sep 14, 2003 11:15 am    Post subject: Reply with quote

Hello there!
Can anyone confirm the usability of the nvidia-kernel module?
I havce tried the 4496-r3 release but my machine hangs with a black screen when I execute startx.
Ctrl+alt+del works and the machine reboots fine.
Any clues?
_________________
linux user 65882
Back to top
View user's profile Send private message
Lovechild
Advocate
Advocate


Joined: 17 May 2002
Posts: 2858
Location: Århus, Denmark

PostPosted: Sun Sep 14, 2003 11:32 am    Post subject: Reply with quote

zypher wrote:
Hello there!
Can anyone confirm the usability of the nvidia-kernel module?
I havce tried the 4496-r3 release but my machine hangs with a black screen when I execute startx.
Ctrl+alt+del works and the machine reboots fine.
Any clues?


nvidia ROCKS on my machine... the issue is probably with X let's see the XFree log.
Back to top
View user's profile Send private message
zypher
Retired Dev
Retired Dev


Joined: 10 Jun 2002
Posts: 416
Location: Cologne, ger.

PostPosted: Sun Sep 14, 2003 2:35 pm    Post subject: Reply with quote

The XFree log ends with:
Code:
(II) NVIDIA(0): Virtual screen size determined to be 1280 x 1024
(==) NVIDIA(0): DPI set to (75, 75)
(II) Loading sub module "fb"
(II) LoadModule: "fb"
(II) Loading /usr/X11R6/lib/modules/libfb.a
(II) Module fb: vendor="The XFree86 Project"
        compiled for 4.3.99.11, module version = 1.0.0
        ABI class: XFree86 ANSI C Emulation, version 0.2
(II) Loading sub module "ramdac"
(II) LoadModule: "ramdac"
(II) Loading /usr/X11R6/lib/modules/libramdac.a
(II) Module ramdac: vendor="The XFree86 Project"
        compiled for 4.3.99.11, module version = 0.1.0
        ABI class: XFree86 Video Driver, version 0.7
(II) do I need RAC?  No, I don't.
(II) resource ranges after preInit:
        [0] 0   0       0xc0000000 - 0xc7ffffff (0x8000000) MX[B]
        [1] 0   0       0xce000000 - 0xceffffff (0x1000000) MX[B]
        [2] -1  0       0xffe00000 - 0xffffffff (0x200000) MX[B](B)
        [3] -1  0       0x00100000 - 0x3fffffff (0x3ff00000) MX[B]E(B)
        [4] -1  0       0x000f0000 - 0x000fffff (0x10000) MX[B]
        [5] -1  0       0x000c0000 - 0x000effff (0x30000) MX[B]
        [6] -1  0       0x00000000 - 0x0009ffff (0xa0000) MX[B]
        [7] -1  0       0xcdbff000 - 0xcdbfffff (0x1000) MX[B]
        [8] -1  0       0xcfe00000 - 0xcfefffff (0x100000) MX[B]
        [9] -1  0       0xcfffa000 - 0xcfffafff (0x1000) MX[B]
        [10] -1 0       0xcfffb000 - 0xcfffbfff (0x1000) MX[B]
        [11] -1 0       0xcffff000 - 0xcfffffff (0x1000) MX[B]
        [12] -1 0       0xcfffe000 - 0xcfffefff (0x1000) MX[B]
        [13] -1 0       0xd0000000 - 0xcfffffff (0x0) MX[B]O
        [14] -1 0       0xcfcf0000 - 0xcfcfffff (0x10000) MX[B](B)
        [15] -1 0       0xc0000000 - 0xc7ffffff (0x8000000) MX[B](B)
        [16] -1 0       0xce000000 - 0xceffffff (0x1000000) MX[B](B)
        [17] -1 0       0xcdbfe000 - 0xcdbfefff (0x1000) MX[B](B)
        [18] 0  0       0x000a0000 - 0x000affff (0x10000) MS[B](OprD)
        [19] 0  0       0x000b0000 - 0x000b7fff (0x8000) MS[B](OprD)
        [20] 0  0       0x000b8000 - 0x000bffff (0x8000) MS[B](OprD)
        [21] -1 0       0x0000ffff - 0x0000ffff (0x1) IX[B]
        [22] -1 0       0x00000000 - 0x000000ff (0x100) IX[B]
        [23] -1 0       0x0000d400 - 0x0000d4ff (0x100) IX[B]
        [24] -1 0       0x0000d800 - 0x0000d83f (0x40) IX[B]
        [25] -1 0       0x0000dc00 - 0x0000dcff (0x100) IX[B]
        [26] -1 0       0x0000ff00 - 0x0000ff0f (0x10) IX[B]
        [27] 0  0       0x000003b0 - 0x000003bb (0xc) IS[B](OprU)
        [28] 0  0       0x000003c0 - 0x000003df (0x20) IS[B](OprU)
(II) NVIDIA(0): AGP 4X successfully initialized
(II) NVIDIA(0): Setting mode "1280x1024"

While a normal log proceeds:
Code:

[26] -1 0       0x0000ff00 - 0x0000ff0f (0x10) IX[B]
[27] 0  0       0x000003b0 - 0x000003bb (0xc) IS[B](OprU)
[28] 0  0       0x000003c0 - 0x000003df (0x20) IS[B](OprU)
(II) NVIDIA(0): AGP 4X successfully initialized
(II) NVIDIA(0): Setting mode "1280x1024"
(II) NVIDIA(0): Using the NVIDIA 2D acceleration architecture
(==) NVIDIA(0): Backing store disabled
(==) NVIDIA(0): Silken mouse enabled
(II) NVIDIA(0): v4l[/dev/video0]: using hw video scaling [YUY2].
(II) Loading extension NV-GLX
(II) Loading extension NV-CONTROL
(==) RandR enabled
(II) Initializing built-in extension MIT-SHM
(II) Initializing built-in extension XInputExtension
(II) Initializing built-in extension XTEST
(II) Initializing built-in extension XKEYBOARD
(II) Initializing built-in extension LBX
(II) Initializing built-in extension XC-APPGROUP
(II) Initializing built-in extension SECURITY
(II) Initializing built-in extension XINERAMA
(II) Initializing built-in extension XFree86-Bigfont
(II) Initializing built-in extension RENDER
(II) Initializing built-in extension RANDR
(II) [GLX]: Initializing GLX extension
(**) Option "Protocol" "PS/2"
(**) Mouse0: Protocol: "PS/2"
(**) Option "CorePointer"
(**) Mouse0: Core Pointer
(**) Option "Device" "/dev/psaux"
(**) Mouse0: Emulate3Buttons, Emulate3Timeout: 50
(==) Mouse0: Buttons: 3
(**) Option "SampleRate" "200"
(**) Mouse0: SampleRate: 200
(**) Option "Resolution" "1800"
(**) Mouse0: Resolution: 1800
(II) Keyboard "Keyboard0" handled by legacy driver
(II) XINPUT: Adding extended input device "Mouse0" (type: MOUSE)
(II) XINPUT: Adding extended input device "NVIDIA Event Handler" (type: Other)
(II) Mouse0: ps2EnableDataReporting: succeeded
Could not init font path element /usr/X11R6/lib/X11/fonts/Speedo/, removing from list!
Could not init font path element /usr/X11R6/lib/X11/fonts/Type1/, removing from list!
Could not init font path element /usr/X11R6/lib/X11/fonts/CID/, removing from list!
GetModeLine - scrn: 0 clock: 135000
GetModeLine - hdsp: 1280 hbeg: 1296 hend: 1440 httl: 1688
              vdsp: 1024 vbeg: 1025 vend: 1028 vttl: 1066 flags: 5
(II) NVIDIA(0): v4l: memPhysBase=0xc0000000


So I should try to force AGP2x?
How do I do this?
_________________
linux user 65882
Back to top
View user's profile Send private message
Safrax
Guru
Guru


Joined: 23 Apr 2002
Posts: 422

PostPosted: Sun Sep 14, 2003 3:13 pm    Post subject: Reply with quote

Are you using nvidia's agp driver or the kernel's? I'm using the kernel driver and haven't had any problems with it.
Back to top
View user's profile Send private message
zypher
Retired Dev
Retired Dev


Joined: 10 Jun 2002
Posts: 416
Location: Cologne, ger.

PostPosted: Sun Sep 14, 2003 4:23 pm    Post subject: Reply with quote

Hmmm actually I thought I'd use the kernls agpgart.
lsmod shows it.
But when I set Option NVAGP "2" the log says "cannot determine agp status".
With 2.4.20-gaming3 everything's fine...
Maybe I try vanilla-2.6.0
_________________
linux user 65882
Back to top
View user's profile Send private message
neenee
Veteran
Veteran


Joined: 20 Jul 2003
Posts: 1786

PostPosted: Sun Sep 14, 2003 4:48 pm    Post subject: Reply with quote

i left agpgart out of the kernel with my nvidia card,
compiled nvidia-glx + nvidia-kernel and set NvAGP
to 1 in my /etc/X11/XF86Config.
_________________
proud to be a scout and a chronic penguin hugger
Legion of Lore - site
Back to top
View user's profile Send private message
zypher
Retired Dev
Retired Dev


Joined: 10 Jun 2002
Posts: 416
Location: Cologne, ger.

PostPosted: Sun Sep 14, 2003 4:51 pm    Post subject: Reply with quote

Mom, I'll try this...
_________________
linux user 65882
Back to top
View user's profile Send private message
Lovechild
Advocate
Advocate


Joined: 17 May 2002
Posts: 2858
Location: Århus, Denmark

PostPosted: Sun Sep 14, 2003 5:09 pm    Post subject: Reply with quote

zypher wrote:
Mom, I'll try this...


Mom ?
Back to top
View user's profile Send private message
zypher
Retired Dev
Retired Dev


Joined: 10 Jun 2002
Posts: 416
Location: Cologne, ger.

PostPosted: Sun Sep 14, 2003 5:21 pm    Post subject: Reply with quote

:)
Sorry, that's short for "one moment please".
I think I mixed a bit of german into my post...

Well I tried startx with NVAGP "1" but it was the same result as my first example.
(I have agpgart as a module)

ATM I have no time to try vanilla-2.6.0 so I think it'll have to wait.
_________________
linux user 65882
Back to top
View user's profile Send private message
zypher
Retired Dev
Retired Dev


Joined: 10 Jun 2002
Posts: 416
Location: Cologne, ger.

PostPosted: Sun Sep 14, 2003 5:29 pm    Post subject: Reply with quote

BTW, lovechild, which nvidia-kernel ebuild works for you?
_________________
linux user 65882
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Off the Wall All times are GMT
Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next
Page 4 of 7

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum