Kernel 2.5.6 Interactive performance

Kernel 2.5.6 Interactive performance

[Dan Mann]

Just compiled 2.5.6 for my workstation at home, where I do 
email/mulitmedia/games, and I wanted to let the everyone that the Interactive 
performance seems fantastic.  I don't know if it's me, the kernel, or my 
hardware, but this seems like the best version I've tested so far.  Heavy 
disk activity (Ide is all I have) such as untarring a kernel tarball and then 
doing a copy either locally or across to a smb server doesn't cause any 
dropouts in xine.  Mouse stays responsive as well.  Kernel compile also has 
the same effect.

Basically, with this kerenl under my loads, the box never feels "loaded".  It 
used to in the past, especially with disk access.  

I don't normally do any throughput tests, but kernel compile times and 
copying/untarring doesn't seem to have slowed down any fwiw.

Machine now feels more responsive than windows 2000 pro machine at work.  


Great work guys.

Dan



The following is some specific info about the hardware and kernel config:

*** Begin Kernel Config ***DMESG below***

#
# Automatically generated by make menuconfig: don't edit
#
CONFIG_X86=y
CONFIG_ISA=y
# CONFIG_SBUS is not set
CONFIG_UID16=y

#
# Code maturity level options
#
CONFIG_EXPERIMENTAL=y

#
# General setup
#
CONFIG_NET=y
CONFIG_SYSVIPC=y
# CONFIG_BSD_PROCESS_ACCT is not set
CONFIG_SYSCTL=y

#
# Loadable module support
#
CONFIG_MODULES=y
CONFIG_MODVERSIONS=y
CONFIG_KMOD=y

#
# Processor type and features
#

CONFIG_MK7=y

CONFIG_X86_WP_WORKS_OK=y
CONFIG_X86_INVLPG=y
CONFIG_X86_CMPXCHG=y
CONFIG_X86_XADD=y
CONFIG_X86_BSWAP=y
CONFIG_X86_POPAD_OK=y

CONFIG_RWSEM_XCHGADD_ALGORITHM=y
CONFIG_X86_L1_CACHE_SHIFT=6
CONFIG_X86_TSC=y
CONFIG_X86_GOOD_APIC=y
CONFIG_X86_USE_3DNOW=y
CONFIG_X86_PGE=y
CONFIG_X86_USE_PPRO_CHECKSUM=y

CONFIG_NOHIGHMEM=y

CONFIG_MTRR=y
# CONFIG_SMP is not set
CONFIG_PREEMPT=y
CONFIG_X86_UP_APIC=y
CONFIG_X86_UP_IOAPIC=y
CONFIG_X86_LOCAL_APIC=y
CONFIG_X86_IO_APIC=y
CONFIG_HAVE_DEC_LOCK=y

#
# General options
#
CONFIG_PCI=y
# CONFIG_PCI_GOBIOS is not set
# CONFIG_PCI_GODIRECT is not set
CONFIG_PCI_GOANY=y
CONFIG_PCI_BIOS=y
CONFIG_PCI_DIRECT=y
CONFIG_PCI_NAMES=y
# CONFIG_EISA is not set
# CONFIG_MCA is not set
CONFIG_HOTPLUG=y

#
# PCMCIA/CardBus support
#
# CONFIG_PCMCIA is not set
# CONFIG_PCMCIA_SA1100 is not set


CONFIG_KCORE_ELF=y
# CONFIG_KCORE_AOUT is not set
# CONFIG_BINFMT_AOUT is not set
CONFIG_BINFMT_ELF=y
CONFIG_BINFMT_MISC=y
# CONFIG_PM is not set
# CONFIG_ACPI is not set
# CONFIG_APM is not set

#
# Plug and Play configuration
#
CONFIG_PNP=y
# CONFIG_ISAPNP is not set
CONFIG_PNPBIOS=y

#
# Block devices
#
CONFIG_BLK_DEV_FD=y


#
# Networking options
#
CONFIG_PACKET=y

CONFIG_UNIX=y
CONFIG_INET=y
CONFIG_IP_MULTICAST=y

#
# ATA/IDE/MFM/RLL support
#
CONFIG_IDE=y

#
# IDE, ATA and ATAPI Block devices
#
CONFIG_BLK_DEV_IDE=y

CONFIG_BLK_DEV_IDEDISK=y
CONFIG_IDEDISK_MULTI_MODE=y

CONFIG_BLK_DEV_IDECD=y

CONFIG_BLK_DEV_IDEPCI=y
CONFIG_IDEPCI_SHARE_IRQ=y
CONFIG_BLK_DEV_IDEDMA_PCI=y

CONFIG_IDEDMA_PCI_AUTO=y
CONFIG_IDEDMA_ONLYDISK=y
CONFIG_BLK_DEV_IDEDMA=y

CONFIG_BLK_DEV_ADMA=y

CONFIG_BLK_DEV_SIS5513=y

CONFIG_IDEDMA_AUTO=y

CONFIG_BLK_DEV_IDE_MODES=y



#
# Network device support
#
CONFIG_NETDEVICES=y


CONFIG_DUMMY=m


#
# Ethernet (10 or 100Mbit)
#
CONFIG_NET_ETHERNET=y

CONFIG_NET_VENDOR_3COM=y

CONFIG_VORTEX=y


#
# Input device support
#
CONFIG_INPUT=y
CONFIG_INPUT_KEYBDEV=y
CONFIG_INPUT_MOUSEDEV=y
CONFIG_INPUT_MOUSEDEV_SCREEN_X=1024
CONFIG_INPUT_MOUSEDEV_SCREEN_Y=768

CONFIG_SOUND_GAMEPORT=y


#
# Character devices
#
CONFIG_VT=y
CONFIG_VT_CONSOLE=y

CONFIG_UNIX98_PTYS=y
CONFIG_UNIX98_PTY_COUNT=256


#
# Mice
#
# CONFIG_BUSMOUSE is not set
CONFIG_MOUSE=y
CONFIG_PSMOUSE=y

CONFIG_AGP=y

CONFIG_DRM=y
CONFIG_DRM_TDFX=y



#
# File systems
#

CONFIG_AUTOFS4_FS=y

CONFIG_EXT3_FS=y
CONFIG_JBD=y

CONFIG_TMPFS=y
CONFIG_RAMFS=y
CONFIG_ISO9660_FS=y

CONFIG_PROC_FS=y

CONFIG_DEVPTS_FS=y

CONFIG_EXT2_FS=y


#
# Network File Systems
#

CONFIG_SMB_FS=y
CONFIG_SMB_NLS_DEFAULT=y
CONFIG_SMB_NLS_REMOTE="cp437"


#
# Partition Types
#
t
CONFIG_MSDOS_PARTITION=y
CONFIG_SMB_NLS=y
CONFIG_NLS=y

#
# Native Language Support
#
CONFIG_NLS_DEFAULT="iso8859-1"
CONFIG_NLS_CODEPAGE_437=y


#
# Console drivers
#
CONFIG_VGA_CONSOLE=y




#
# Sound
#
CONFIG_SOUND=y

#
# Open Sound System
#


#
# Advanced Linux Sound Architecture
#


#
# Generic devices
#


#
# ISA devices
#


#
# PCI devices

CONFIG_SND_CMIPCI=y


#
# USB support
#
CONFIG_USB=y

CONFIG_USB_OHCI=y

CONFIG_USB_HID=y

********Begin DMESG********

Mar  9 00:43:46 cronos syslogd 1.4.1: restart.
Mar  9 00:43:47 cronos kernel: klogd 1.4.1, log source = /proc/kmsg started.
Mar  9 00:43:47 cronos kernel: Inspecting /boot/System.map-2.5.6
Mar  9 00:43:47 cronos kernel: Loaded 17190 symbols from 
/boot/System.map-2.5.6.
Mar  9 00:43:47 cronos kernel: Symbols match kernel version 2.5.6.
Mar  9 00:43:47 cronos kernel: No module symbols loaded.
Mar  9 00:43:47 cronos kernel: Linux version 2.5.6 (root@cronos) (gcc version 
2.96 20000731 (Mandrake Linux 8.2 2.96-0.74mdk)) #1 Fri Mar 8 23:29:34 EST 
2002
Mar  9 00:43:47 cronos kernel: BIOS-provided physical RAM map:
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 0000000000000000 - 
000000000009fc00 (usable)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 000000000009fc00 - 
00000000000a0000 (reserved)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 00000000000f0000 - 
0000000000100000 (reserved)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 0000000000100000 - 
000000000c000000 (usable)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 00000000fec00000 - 
00000000fec01000 (reserved)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 00000000fee00000 - 
00000000fee01000 (reserved)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 00000000ffee0000 - 
00000000fff00000 (reserved)
Mar  9 00:43:47 cronos kernel:  BIOS-e820: 00000000fffc0000 - 
0000000100000000 (reserved)
Mar  9 00:43:47 cronos kernel: 192MB LOWMEM available.
Mar  9 00:43:47 cronos kernel: On node 0 totalpages: 49152
Mar  9 00:43:47 cronos kernel: zone(0): 4096 pages.
Mar  9 00:43:47 cronos kernel: zone(1): 45056 pages.
Mar  9 00:43:47 cronos kernel: zone(2): 0 pages.
Mar  9 00:43:47 cronos kernel: Kernel command line: ro root=/dev/hda5
Mar  9 00:43:47 cronos kernel: Local APIC disabled by BIOS -- reenabling.
Mar  9 00:43:47 cronos kernel: Found and enabled local APIC!
Mar  9 00:43:47 cronos kernel: Initializing CPU#0
Mar  9 00:43:47 cronos kernel: Detected 945.987 MHz processor.
Mar  9 00:43:47 cronos kernel: Console: colour VGA+ 80x25
Mar  9 00:43:47 cronos kernel: Calibrating delay loop... 1887.43 BogoMIPS
Mar  9 00:43:47 cronos kernel: Memory: 191972k/196608k available (1336k 
kernel code, 4248k reserved, 333k data, 212k init, 0k highmem)
Mar  9 00:43:47 cronos kernel: Dentry-cache hash table entries: 32768 (order: 
6, 262144 bytes)
Mar  9 00:43:47 cronos kernel: Inode-cache hash table entries: 16384 (order: 
5, 131072 bytes)
Mar  9 00:43:47 cronos kernel: Mount-cache hash table entries: 512 (order: 0, 
4096 bytes)
Mar  9 00:43:47 cronos kernel: Buffer-cache hash table entries: 16384 (order: 
4, 65536 bytes)
Mar  9 00:43:47 cronos kernel: Page-cache hash table entries: 65536 (order: 
6, 262144 bytes)
Mar  9 00:43:47 cronos kernel: CPU: Before vendor init, caps: 0183fbff 
c1c7fbff 00000000, vendor = 2
Mar  9 00:43:47 cronos kernel: CPU: L1 I Cache: 64K (64 bytes/line), D cache 
64K (64 bytes/line)
Mar  9 00:43:47 cronos kernel: CPU: L2 Cache: 256K (64 bytes/line)
Mar  9 00:43:47 cronos kernel: CPU: After vendor init, caps: 0183fbff 
c1c7fbff 00000000 00000000
Mar  9 00:43:47 cronos kernel: Intel machine check architecture supported.
Mar  9 00:43:47 cronos kernel: Intel machine check reporting enabled on CPU#0.
Mar  9 00:43:47 cronos kernel: CPU:     After generic, caps: 0183fbff 
c1c7fbff 00000000 00000000
Mar  9 00:43:47 cronos kernel: CPU:             Common caps: 0183fbff 
c1c7fbff 00000000 00000000
Mar  9 00:43:47 cronos kernel: CPU: AMD Athlon(tm) Processor stepping 02
Mar  9 00:43:47 cronos kernel: Enabling fast FPU save and restore... done.
Mar  9 00:43:47 cronos kernel: Checking 'hlt' instruction... OK.
Mar  9 00:43:47 cronos kernel: POSIX conformance testing by UNIFIX
Mar  9 00:43:47 cronos kernel: enabled ExtINT on CPU#0
Mar  9 00:43:47 cronos kernel: ESR value before enabling vector: 00000000
Mar  9 00:43:47 cronos kernel: ESR value after enabling vector: 00000000
Mar  9 00:43:47 cronos kernel: Using local APIC timer interrupts.
Mar  9 00:43:47 cronos kernel: calibrating APIC timer ...
Mar  9 00:43:47 cronos kernel: ..... CPU clock speed is 946.0004 MHz.
Mar  9 00:43:47 cronos kernel: ..... host bus clock speed is 199.1579 MHz.
Mar  9 00:43:47 cronos kernel: cpu: 0, clocks: 1991579, slice: 995789
Mar  9 00:43:47 cronos kernel: 
CPU0<T0:1991568,T1:995776,D:3,S:995789,C:1991579>
Mar  9 00:43:47 cronos kernel: mtrr: v1.40 (20010327) Richard Gooch 
(rgooch@atnf.csiro.au)
Mar  9 00:43:47 cronos kernel: mtrr: detected mtrr type: Intel
Mar  9 00:43:47 cronos kernel: Linux NET4.0 for Linux 2.4
Mar  9 00:43:47 cronos kernel: Based upon Swansea University Computer Society 
NET3.039
Mar  9 00:43:47 cronos kernel: Initializing RT netlink socket
Mar  9 00:43:47 cronos kernel: PCI: PCI BIOS revision 2.10 entry at 0xfdb01, 
last bus=1
Mar  9 00:43:47 cronos kernel: PCI: Using configuration type 1
Mar  9 00:43:47 cronos kernel: PCI: Probing PCI hardware
Mar  9 00:43:47 cronos kernel: PCI: Using IRQ router SIS [1039/0008] at 
00:02.0
Mar  9 00:43:47 cronos kernel: PnPBIOS: Found PnP BIOS installation structure 
at 0xc00f7210.
Mar  9 00:43:47 cronos kernel: PnPBIOS: PnP BIOS version 1.0, entry 
0xf0000:0x5cc7, dseg 0xf0000.
Mar  9 00:43:47 cronos kernel: PnPBIOS: 13 nodes reported by PnP BIOS; 13 
recorded by driver.
Mar  9 00:43:47 cronos kernel: Starting kswapd
Mar  9 00:43:47 cronos kernel: BIO: pool of 256 setup, 14Kb (56 bytes/bio)
Mar  9 00:43:47 cronos kernel: biovec: init pool 0, 1 entries, 12 bytes
Mar  9 00:43:47 cronos kernel: biovec: init pool 1, 4 entries, 48 bytes
Mar  9 00:43:47 cronos kernel: biovec: init pool 2, 16 entries, 192 bytes
Mar  9 00:43:47 cronos kernel: biovec: init pool 3, 64 entries, 768 bytes
Mar  9 00:43:47 cronos kernel: biovec: init pool 4, 128 entries, 1536 bytes
Mar  9 00:43:47 cronos kernel: biovec: init pool 5, 256 entries, 3072 bytes
Mar  9 00:43:47 cronos kernel: Journalled Block Device driver loaded
Mar  9 00:43:47 cronos kernel: pty: 256 Unix98 ptys configured
Mar  9 00:43:47 cronos kernel: block: 256 slots per queue, batch=32
Mar  9 00:43:47 cronos kernel: Floppy drive(s): fd0 is 1.44M
Mar  9 00:43:47 cronos kernel: FDC 0 is a post-1991 82077
Mar  9 00:43:47 cronos kernel: PCI: Found IRQ 11 for device 00:0f.0
Mar  9 00:43:47 cronos kernel: PCI: Sharing IRQ 11 with 00:02.2
Mar  9 00:43:47 cronos kernel: 3c59x: Donald Becker and others. 
www.scyld.com/network/vortex.html
Mar  9 00:43:47 cronos kernel: 00:0f.0: 3Com PCI 3c905B Cyclone 100baseTx at 
0xd800. Vers LK1.1.16
Mar  9 00:43:47 cronos kernel: Linux agpgart interface v0.99 (c) Jeff Hartmann
Mar  9 00:43:47 cronos kernel: agpgart: Maximum main memory to use for agp 
memory: 150M
Mar  9 00:43:47 cronos kernel: agpgart: Detected SiS 735 chipset
Mar  9 00:43:47 cronos kernel: agpgart: AGP aperture is 64M @ 0xd0000000
Mar  9 00:43:47 cronos kernel: [drm] Initialized tdfx 1.0.0 20010216 on minor 
0
Mar  9 00:43:47 cronos kernel: Uniform Multi-Platform E-IDE driver ver.:7.0.0
Mar  9 00:43:47 cronos kernel: ide: system bus speed 33MHz
Mar  9 00:43:47 cronos kernel: Silicon Integrated Systems [SiS] 5513 [IDE]: 
IDE controller on PCI slot 00:02.5
Mar  9 00:43:47 cronos kernel: Silicon Integrated Systems [SiS] 5513 [IDE]: 
chipset revision 208
Mar  9 00:43:47 cronos kernel: Silicon Integrated Systems [SiS] 5513 [IDE]: 
not 100%% native mode: will probe irqs later
Mar  9 00:43:47 cronos kernel: SiS735
Mar  9 00:43:47 cronos kernel:     ide0: BM-DMA at 0xff00-0xff07, BIOS 
settings: hda:DMA, hdb:DMA
Mar  9 00:43:47 cronos kernel:     ide1: BM-DMA at 0xff08-0xff0f, BIOS 
settings: hdc:DMA, hdd:DMA
Mar  9 00:43:47 cronos kernel: hda: Maxtor 52049U4, ATA DISK drive
Mar  9 00:43:47 cronos kernel: hdc: Hewlett-Packard CD-Writer Plus 9100, 
ATAPI CD/DVD-ROM drive
Mar  9 00:43:47 cronos kernel: ide0 at 0x1f0-0x1f7,0x3f6 on irq 14
Mar  9 00:43:47 cronos kernel: ide1 at 0x170-0x177,0x376 on irq 15
Mar  9 00:43:47 cronos kernel: blk: queue c030636c, I/O limit 4095Mb (mask 
0xffffffff)
Mar  9 00:43:47 cronos kernel: hda: 40020624 sectors (20491 MB) w/2048KiB 
Cache, CHS=39703/16/63, UDMA(66)
Mar  9 00:43:47 cronos kernel: hdc: ATAPI 32X CD-ROM CD-R/RW drive, 4096kB 
Cache, UDMA(33)
Mar  9 00:43:47 cronos kernel: Uniform CD-ROM driver Revision: 3.12
Mar  9 00:43:47 cronos kernel: Partition check:
Mar  9 00:43:47 cronos kernel:  hda: [PTBL] [2491/255/63] hda1 hda2 hda3 < 
hda5 hda6 hda7 hda8 >
Mar  9 00:43:47 cronos kernel: Advanced Linux Sound Architecture Driver 
Version 0.9.0beta12 (Wed Mar 06 07:56:20 2002 UTC).
Mar  9 00:43:47 cronos kernel: kmod: failed to exec /sbin/modprobe -s -k 
snd-card-0, errno = 2
Mar  9 00:43:47 cronos kernel: PCI: Found IRQ 5 for device 00:0d.0
Mar  9 00:43:47 cronos kernel: cmipci: no OPL device at 0x388
Mar  9 00:43:47 cronos kernel: ALSA device list:
Mar  9 00:43:47 cronos kernel:   #0: C-Media PCI CMI8738 (model 37) at 
0xdc00, irq 5
Mar  9 00:43:47 cronos kernel: usb.c: registered new driver hub
Mar  9 00:43:47 cronos kernel: SiS router pirq escape (99)
Mar  9 00:43:47 cronos kernel: SiS router pirq escape (99)
Mar  9 00:43:47 cronos kernel: usb-ohci.c: USB OHCI at membase 0xcc816000, 
IRQ 10
Mar  9 00:43:47 cronos kernel: usb-ohci.c: usb-00:02.3, Silicon Integrated 
Systems [SiS] 7001 (#2)
Mar  9 00:43:47 cronos kernel: hcd.c: new USB bus registered, assigned bus 
number 1
Mar  9 00:43:47 cronos kernel: hub.c: USB hub found at /
Mar  9 00:43:47 cronos kernel: hub.c: 3 ports detected
Mar  9 00:43:47 cronos kernel: PCI: Found IRQ 11 for device 00:02.2
Mar  9 00:43:47 cronos kernel: PCI: Sharing IRQ 11 with 00:0f.0
Mar  9 00:43:47 cronos kernel: usb-ohci.c: USB OHCI at membase 0xcc818000, 
IRQ 11
Mar  9 00:43:47 cronos kernel: usb-ohci.c: usb-00:02.2, Silicon Integrated 
Systems [SiS] 7001
Mar  9 00:43:47 cronos kernel: hcd.c: new USB bus registered, assigned bus 
number 2
Mar  9 00:43:47 cronos kernel: hub.c: USB hub found at /
Mar  9 00:43:47 cronos kernel: hub.c: 3 ports detected
Mar  9 00:43:47 cronos kernel: usb.c: registered new driver hid
Mar  9 00:43:47 cronos kernel: hid-core.c: v1.31:USB HID core driver
Mar  9 00:43:47 cronos kernel: mice: PS/2 mouse device common for all mice
Mar  9 00:43:47 cronos kernel: NET4: Linux TCP/IP 1.0 for NET4.0
Mar  9 00:43:47 cronos kernel: IP Protocols: ICMP, UDP, TCP, IGMP
Mar  9 00:43:47 cronos kernel: IP: routing cache hash table of 2048 buckets, 
16Kbytes
Mar  9 00:43:47 cronos kernel: TCP: Hash tables configured (established 16384 
bind 16384)
Mar  9 00:43:47 cronos kernel: NET4: Unix domain sockets 1.0/SMP for Linux 
NET4.0.
Mar  9 00:43:47 cronos kernel: kjournald starting.  Commit interval 5 seconds
Mar  9 00:43:47 cronos kernel: EXT3-fs: mounted filesystem with ordered data 
mode.
Mar  9 00:43:47 cronos kernel: VFS: Mounted root (ext3 filesystem) readonly.
Mar  9 00:43:47 cronos kernel: Freeing unused kernel memory: 212k freed
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 2
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=2 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 3
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=3 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 4
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=4 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 5
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=5 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 6
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=6 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 7
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=7 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 8
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=8 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 9
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=9 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 10
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=10 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 11
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=11 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 12
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=12 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 13
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=13 
(error=-110)
Mar  9 00:43:47 cronos kernel: EXT3 FS 2.4-0.9.16, 02 Dec 2001 on ide0(3,5), 
internal journal
Mar  9 00:43:47 cronos kernel: Adding Swap: 192740k swap-space (priority -1)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 14
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=14 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 15
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=15 
(error=-110)
Mar  9 00:43:47 cronos random: Initializing random number generator:  
succeeded
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 16
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=16 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 17
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=17 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 18
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=18 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 19
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=19 
(error=-110)
Mar  9 00:43:47 cronos kernel: kjournald starting.  Commit interval 5 seconds
Mar  9 00:43:47 cronos kernel: EXT3 FS 2.4-0.9.16, 02 Dec 2001 on ide0(3,1), 
internal journal
Mar  9 00:43:47 cronos kernel: EXT3-fs: mounted filesystem with ordered data 
mode.
Mar  9 00:43:47 cronos kernel: kjournald starting.  Commit interval 5 seconds
Mar  9 00:43:47 cronos kernel: EXT3 FS 2.4-0.9.16, 02 Dec 2001 on ide0(3,2), 
internal journal
Mar  9 00:43:47 cronos kernel: EXT3-fs: mounted filesystem with ordered data 
mode.
Mar  9 00:43:47 cronos kernel: kjournald starting.  Commit interval 5 seconds
Mar  9 00:43:47 cronos kernel: EXT3 FS 2.4-0.9.16, 02 Dec 2001 on ide0(3,6), 
internal journal
Mar  9 00:43:47 cronos kernel: EXT3-fs: mounted filesystem with ordered data 
mode.
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 20
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=20 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 21
Mar  9 00:43:47 cronos kernel: usb.c: USB device not accepting new address=21 
(error=-110)
Mar  9 00:43:47 cronos kernel: hub.c: new USB device on bus 1 path /1, 
assigned address 22
Mar  9 00:43:47 cronos kernel: hid-core.c: ctrl urb status -32 received
Mar  9 00:43:47 cronos kernel: usb_control/bulk_msg: timeout
Mar  9 00:43:47 cronos kernel: input.c: calling /sbin/hotplug input [HOME=/ 
PATH=/sbin:/bin:/usr/sbin:/usr/bin ACTION=add PRODUCT=3/461/4d09/10 
NAME=0461:4d09]
Mar  9 00:43:47 cronos kernel: input: USB HID v1.00 Mouse [0461:4d09] on 
usb1:1

Re: Kernel 2.5.6 Interactive performance

[Dieter Nützel]

On Saturday, 9. März 2002 18:55:00, Dan Mann wrote:
[-]
>Machine now feels more responsive than windows 2000 pro machine at work.  
>
> Great work guys.

It's due to preemption and Ingo's great O(1)-scheduler.
BIO should help, too but throughput isn't were it should be...;-)

You can get this when you apply preemption+lock-break, O(1) and Andrew 
Morten's low-latency to 2.4.18, too.

-aa (vm_29) deliver additional throughput.
If you are running under KDE you should try 3.0 beta2 or -rc2 (!!!)
It flies then.

Regards,
	Dieter
-- 
Dieter Nützel
Graduate Student, Computer Science

University of Hamburg
Department of Computer Science
@home: Dieter.Nuetzel@hamburg.de

RE: Kernel 2.5.6 Interactive performance

[Charles Heselton]

How would you implement these thing?  I'm not on the same technical level
that you guys are, and when/if things are out of context, I don't follow.
Can you help?

Charles Heselton
Network Installer
Staffing Alternatives, Inc.
619.261.6866
charles_heselton@hotmail.com <mailto:charles_heselton@hotmail.com>




-----Original Message-----
From: linux-kernel-owner@vger.kernel.org
[mailto:linux-kernel-owner@vger.kernel.org]On Behalf Of Dieter Nützel
Sent: Saturday, March 09, 2002 1156
To: Dan Mann
Cc: Linux Kernel List
Subject: Re: Kernel 2.5.6 Interactive performance


On Saturday, 9. März 2002 18:55:00, Dan Mann wrote:
[-]
>Machine now feels more responsive than windows 2000 pro machine at work.
>
> Great work guys.

It's due to preemption and Ingo's great O(1)-scheduler.
BIO should help, too but throughput isn't were it should be...;-)

You can get this when you apply preemption+lock-break, O(1) and Andrew
Morten's low-latency to 2.4.18, too.

-aa (vm_29) deliver additional throughput.
If you are running under KDE you should try 3.0 beta2 or -rc2 (!!!)
It flies then.

Regards,
	Dieter
--
Dieter Nützel
Graduate Student, Computer Science

University of Hamburg
Department of Computer Science
@home: Dieter.Nuetzel@hamburg.de

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: Kernel 2.5.6 Interactive performance

[Dieter Nützel]

On Sonntag, 10. März 2002 02:00:02, Charles Heselton wrote:
> How would you implement these thing?  I'm not on the same technical level
> that you guys are, and when/if things are out of context, I don't follow.
> Can you help?

If you are somewhat open for "new" (experimental) stuff I can prepare a patch 
on top of 2.4.18 or 2.4.19-pre2 for you.

But Andrea Arcangeli informed me that vm-29 had a deadlock bug in the recent 
fixes for the bh headers. vm-28 is fine or soon to be available vm-30.

It didn't hit me for the last two days.

Regards,
	Dieter

> -----Original Message-----
> From: linux-kernel-owner@vger.kernel.org
> [mailto:linux-kernel-owner@vger.kernel.org]On Behalf Of Dieter Nützel
> Sent: Saturday, March 09, 2002 1156
> To: Dan Mann
> Cc: Linux Kernel List
> Subject: Re: Kernel 2.5.6 Interactive performance
>
>
> On Saturday, 9. März 2002 18:55:00, Dan Mann wrote:
> [-]
>
> >Machine now feels more responsive than windows 2000 pro machine at work.
> >
> > Great work guys.
>
> It's due to preemption and Ingo's great O(1)-scheduler.
> BIO should help, too but throughput isn't were it should be...;-)
>
> You can get this when you apply preemption+lock-break, O(1) and Andrew
> Morten's low-latency to 2.4.18, too.
>
> -aa (vm_29) deliver additional throughput.
> If you are running under KDE you should try 3.0 beta2 or -rc2 (!!!)
> It flies then.
>
> Regards,
> 	Dieter

RE: Kernel 2.5.6 Interactive performance

[Charles Heselton]

That would be great.  I'm currently running 2.4.18.  I'm always up for
things that would help improve performance, even if they are "experimental".

Thanks,
Charles Heselton
Network Installer
Staffing Alternatives, Inc.
619.261.6866
charles_heselton@hotmail.com <mailto:charles_heselton@hotmail.com>




-----Original Message-----
From: Dieter Nützel [mailto:Dieter.Nuetzel@hamburg.de]
Sent: Saturday, March 09, 2002 1712
To: charles-heselton@cox.net; Dan Mann
Cc: Linux Kernel List; J.A. Magallon
Subject: Re: Kernel 2.5.6 Interactive performance


On Sonntag, 10. März 2002 02:00:02, Charles Heselton wrote:
> How would you implement these thing?  I'm not on the same technical level
> that you guys are, and when/if things are out of context, I don't follow.
> Can you help?

If you are somewhat open for "new" (experimental) stuff I can prepare a
patch
on top of 2.4.18 or 2.4.19-pre2 for you.

But Andrea Arcangeli informed me that vm-29 had a deadlock bug in the recent
fixes for the bh headers. vm-28 is fine or soon to be available vm-30.

It didn't hit me for the last two days.

Regards,
	Dieter

> -----Original Message-----
> From: linux-kernel-owner@vger.kernel.org
> [mailto:linux-kernel-owner@vger.kernel.org]On Behalf Of Dieter Nützel
> Sent: Saturday, March 09, 2002 1156
> To: Dan Mann
> Cc: Linux Kernel List
> Subject: Re: Kernel 2.5.6 Interactive performance
>
>
> On Saturday, 9. März 2002 18:55:00, Dan Mann wrote:
> [-]
>
> >Machine now feels more responsive than windows 2000 pro machine at work.
> >
> > Great work guys.
>
> It's due to preemption and Ingo's great O(1)-scheduler.
> BIO should help, too but throughput isn't were it should be...;-)
>
> You can get this when you apply preemption+lock-break, O(1) and Andrew
> Morten's low-latency to 2.4.18, too.
>
> -aa (vm_29) deliver additional throughput.
> If you are running under KDE you should try 3.0 beta2 or -rc2 (!!!)
> It flies then.
>
> Regards,
> 	Dieter

RE: Kernel 2.5.6 Interactive performance

[Robert Love]

On Sat, 2002-03-09 at 20:15, Charles Heselton wrote:

> That would be great.  I'm currently running 2.4.18.  I'm always up for
> things that would help improve performance, even if they are "experimental".

A good base is Alan's tree, available at:

	http://www.kernel.org/pub/linux/kernel/people/alan/linux-2.4/2.4.19/patch-2.4.19-pre2-ac4.gz

which is to be applied on top of 2.4.19-pre2.  It contains the O(1)
scheduler and rmap VM.  If you are interested in preemption, the
preempt-kernel patch is available at:

	http://www.kernel.org/pub/linux/kernel/people/rml/preempt-kernel/v2.4/

The 2.5 tree also has most of these toys, and is a better place for this
development IMO.  Personally, I'd stay away from these all-in-one silly
patches that are floating around these days.  Your safest bet is just
stock 2.4.18 or whatever is latest, although the above addons are all at
varying levels of "stable" and "safe".

	Robert Love

Re: Kernel 2.5.6 Interactive performance

[Mike Fedyk]

On Sat, Mar 09, 2002 at 11:23:48PM -0500, Robert Love wrote:
> On Sat, 2002-03-09 at 20:15, Charles Heselton wrote:
> 
> > That would be great.  I'm currently running 2.4.18.  I'm always up for
> > things that would help improve performance, even if they are "experimental".
> 
> A good base is Alan's tree, available at:
> 
> 	http://www.kernel.org/pub/linux/kernel/people/alan/linux-2.4/2.4.19/patch-2.4.19-pre2-ac4.gz
> 
> which is to be applied on top of 2.4.19-pre2.  It contains the O(1)
> scheduler and rmap VM.  If you are interested in preemption, the
> preempt-kernel patch is available at:
> 
> 	http://www.kernel.org/pub/linux/kernel/people/rml/preempt-kernel/v2.4/
> 
> The 2.5 tree also has most of these toys, and is a better place for this
> development IMO.  Personally, I'd stay away from these all-in-one silly
> patches that are floating around these days.  Your safest bet is just
> stock 2.4.18 or whatever is latest, although the above addons are all at
> varying levels of "stable" and "safe".
> 

Then what do you call -aa and -ac? ;)

These "all-in-one" patches do make it harder to debug specific patches, but
it does create a wider audience for many patches that wouldn't be used
otherwise.

Re: Kernel 2.5.6 Interactive performance

[J Sloan]

Robert Love wrote:

>The 2.5 tree also has most of these toys, and is a better place for this
>development IMO.  Personally, I'd stay away from these all-in-one silly
>patches that are floating around these days.  Your safest bet is just
>stock 2.4.18 or whatever is latest, although the above addons are all at
>varying levels of "stable" and "safe".
>

Just my $.02 -

After futzing around with all the various patches
floating around, I've found the -aa releases to be
a pleasant surprise all around. I generally run -aa
on my home and office workstations, as well as
the web/mail/dns/squid/firewall servers I manage.

I find I get 95% of the benefits of the bleeding
edge, with 5% of the effort - for instance:

untar 2.4.18
apply 2.4.19-pre2 patch
apply 2.4.19-pre2aa1 patch *
configure, compile, boot and enjoy.

* for nvidia drivers, back out xfs and 20_pte-highmem patches

Joe

Re: Kernel 2.5.6 Interactive performance

[Robert Love]

On Sat, 2002-03-09 at 23:38, Mike Fedyk wrote:

> On Sat, Mar 09, 2002 at 11:23:48PM -0500, Robert Love wrote:
> > The 2.5 tree also has most of these toys, and is a better place for this
> > development IMO.  Personally, I'd stay away from these all-in-one silly
> > patches that are floating around these days.  Your safest bet is just
> > stock 2.4.18 or whatever is latest, although the above addons are all at
> > varying levels of "stable" and "safe".
> > 
> 
> Then what do you call -aa and -ac? ;)
> 
> These "all-in-one" patches do make it harder to debug specific patches, but
> it does create a wider audience for many patches that wouldn't be used
> otherwise.

I don't put -aa nor -ac in the same category as what I refer to above. 
Alan and Andrea's trees both contain an intelligent combination of
useful patches, bug fixes, and code from Alan and Andrea themselves.

The plethora of all-in-one every-patch-under-the-sun patchsets don't
fall into the above category, in my opinion.  They just mix various new
feature patches.  They do offer one benefit: much wider exposure for
some potentially very useful patches.  I have found, however, that they
don't help the actual patch authors much since (a) they are mixed in
with many other patches and possibly even erroneously merged and (b) the
bug reports never make it upstream to the actual patch maintainers.

Maybe I'm just annoyed by the even greater signal-to-noise ratio on lkml
:-)

	Robert Love

RE: Kernel 2.5.6 Interactive performance

[Charles Heselton]

Well, unfortunately, you guys are still talking a little above my head.  I
kind of understand what you are saying but not completely.  Are the -aa
and -ac patches?  How do you install/run a patch?  Are they tags to put in
when compiling?  What is VM28-vm30?  All I've done so far is untar the
tarballs from kernel.org (or wherever) and go from there.  Finally started
having success with it, but all this stuff that you guys are talking about
on the development level is a little above me.  Which, BTW, is partly why I
subscribed to the mailing list - to try to learn a little more.  So could
you guys be a little more specific in the explanations?

Thanks,
Charles Heselton
Network Installer
Staffing Alternatives, Inc.
619.261.6866
charles_heselton@hotmail.com <mailto:charles_heselton@hotmail.com>




-----Original Message-----
From: linux-kernel-owner@vger.kernel.org
[mailto:linux-kernel-owner@vger.kernel.org]On Behalf Of Robert Love
Sent: Saturday, March 09, 2002 2206
To: Mike Fedyk
Cc: charles-heselton@cox.net; Dieter N?tzel; Dan Mann; Linux Kernel
List; J.A. Magallon
Subject: Re: Kernel 2.5.6 Interactive performance


On Sat, 2002-03-09 at 23:38, Mike Fedyk wrote:

> On Sat, Mar 09, 2002 at 11:23:48PM -0500, Robert Love wrote:
> > The 2.5 tree also has most of these toys, and is a better place for this
> > development IMO.  Personally, I'd stay away from these all-in-one silly
> > patches that are floating around these days.  Your safest bet is just
> > stock 2.4.18 or whatever is latest, although the above addons are all at
> > varying levels of "stable" and "safe".
> >
>
> Then what do you call -aa and -ac? ;)
>
> These "all-in-one" patches do make it harder to debug specific patches,
but
> it does create a wider audience for many patches that wouldn't be used
> otherwise.

I don't put -aa nor -ac in the same category as what I refer to above.
Alan and Andrea's trees both contain an intelligent combination of
useful patches, bug fixes, and code from Alan and Andrea themselves.

The plethora of all-in-one every-patch-under-the-sun patchsets don't
fall into the above category, in my opinion.  They just mix various new
feature patches.  They do offer one benefit: much wider exposure for
some potentially very useful patches.  I have found, however, that they
don't help the actual patch authors much since (a) they are mixed in
with many other patches and possibly even erroneously merged and (b) the
bug reports never make it upstream to the actual patch maintainers.

Maybe I'm just annoyed by the even greater signal-to-noise ratio on lkml
:-)

	Robert Love

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

RE: Kernel 2.5.6 Interactive performance

[Charles Heselton]

I can understand that.  This just seemed the most readily available
resource.  All of you über-geeks out there...I appreciate your knowledge and
hard work.

Thanks,
Charles Heselton
Network Installer
Staffing Alternatives, Inc.
619.261.6866
charles_heselton@hotmail.com <mailto:charles_heselton@hotmail.com>


-----Original Message-----
From: Hua Zhong [mailto:hzhong@cisco.com]
Sent: Saturday, March 09, 2002 2235
To: charles-heselton@cox.net
Subject: Re: Kernel 2.5.6 Interactive performance


I suggest you subsribe to the kernelnewbies maillist instead. Start from
http://www.kernelnewbies.org.
It's much more useful for a newbie.

LKML is for developing linux kernel, not for tutoring how to use kernel.
Don't expect/ask those hackers
to speak in language easy to understand for everyone.

----- Original Message -----
From: "Charles Heselton" <charles-heselton@cox.net>
To: "Robert Love" <rml@tech9.net>; "Mike Fedyk" <mfedyk@matchmail.com>
Cc: "Dieter N?tzel" <Dieter.Nuetzel@hamburg.de>; "Dan Mann"
<mainlylinux@attbi.com>; "Linux Kernel List" <linux-kernel@vger.kernel.org>;
"J.A. Magallon" <jamagallon@able.es>
Sent: Saturday, March 09, 2002 10:18 PM
Subject: RE: Kernel 2.5.6 Interactive performance


>
> Well, unfortunately, you guys are still talking a little above my head.  I
> kind of understand what you are saying but not completely.  Are the -aa
> and -ac patches?  How do you install/run a patch?  Are they tags to put in
> when compiling?  What is VM28-vm30?  All I've done so far is untar the
> tarballs from kernel.org (or wherever) and go from there.  Finally started
> having success with it, but all this stuff that you guys are talking about
> on the development level is a little above me.  Which, BTW, is partly why
I
> subscribed to the mailing list - to try to learn a little more.  So could
> you guys be a little more specific in the explanations?
>
> Thanks,
> Charles Heselton
> Network Installer
> Staffing Alternatives, Inc.
> 619.261.6866
> charles_heselton@hotmail.com <mailto:charles_heselton@hotmail.com>
>
>
>
>
> -----Original Message-----
> From: linux-kernel-owner@vger.kernel.org
> [mailto:linux-kernel-owner@vger.kernel.org]On Behalf Of Robert Love
> Sent: Saturday, March 09, 2002 2206
> To: Mike Fedyk
> Cc: charles-heselton@cox.net; Dieter N?tzel; Dan Mann; Linux Kernel
> List; J.A. Magallon
> Subject: Re: Kernel 2.5.6 Interactive performance
>
>
> On Sat, 2002-03-09 at 23:38, Mike Fedyk wrote:
>
> > On Sat, Mar 09, 2002 at 11:23:48PM -0500, Robert Love wrote:
> > > The 2.5 tree also has most of these toys, and is a better place for
this
> > > development IMO.  Personally, I'd stay away from these all-in-one
silly
> > > patches that are floating around these days.  Your safest bet is just
> > > stock 2.4.18 or whatever is latest, although the above addons are all
at
> > > varying levels of "stable" and "safe".
> > >
> >
> > Then what do you call -aa and -ac? ;)
> >
> > These "all-in-one" patches do make it harder to debug specific patches,
> but
> > it does create a wider audience for many patches that wouldn't be used
> > otherwise.
>
> I don't put -aa nor -ac in the same category as what I refer to above.
> Alan and Andrea's trees both contain an intelligent combination of
> useful patches, bug fixes, and code from Alan and Andrea themselves.
>
> The plethora of all-in-one every-patch-under-the-sun patchsets don't
> fall into the above category, in my opinion.  They just mix various new
> feature patches.  They do offer one benefit: much wider exposure for
> some potentially very useful patches.  I have found, however, that they
> don't help the actual patch authors much since (a) they are mixed in
> with many other patches and possibly even erroneously merged and (b) the
> bug reports never make it upstream to the actual patch maintainers.
>
> Maybe I'm just annoyed by the even greater signal-to-noise ratio on lkml
> :-)
>
> Robert Love
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/
>
> -
> To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html
> Please read the FAQ at  http://www.tux.org/lkml/

Re: Kernel 2.5.6 Interactive performance

[Andrea Arcangeli]

On Sun, Mar 10, 2002 at 02:11:49AM +0100, Dieter Nützel wrote:
> On Sonntag, 10. März 2002 02:00:02, Charles Heselton wrote:
> > How would you implement these thing?  I'm not on the same technical level
> > that you guys are, and when/if things are out of context, I don't follow.
> > Can you help?
> 
> If you are somewhat open for "new" (experimental) stuff I can prepare a patch 
> on top of 2.4.18 or 2.4.19-pre2 for you.
> 
> But Andrea Arcangeli informed me that vm-29 had a deadlock bug in the recent 
> fixes for the bh headers. vm-28 is fine or soon to be available vm-30.

Confirm. I only wanted to add that only 2.4.19pre2aa1 and vm-29 can
deadlock, all previous -aa kernels and vm-?? patches are rock solid
AFIK. The bug is just fixed, it was a missing UnlockPage. The bug was
introduced while fixing an highmem balancing thing (that nobody ever
reported on production machines so I don't consider such problem a
showstopper, but nevertheless vm-30 will fix both the new pratical and
the old mostly theorical problem).

Andrea