Gentoo Websites Logo
Go to: Gentoo Home Documentation Forums Lists Bugs Planet Store Wiki Get Gentoo!
Bug 142609 - gentoo-sources-2.6.17-r4 b44 drivers has no TX/RX flow control
Summary: gentoo-sources-2.6.17-r4 b44 drivers has no TX/RX flow control
Status: RESOLVED FIXED
Alias: None
Product: Gentoo Linux
Classification: Unclassified
Component: [OLD] Core system (show other bugs)
Hardware: x86 Linux
: High major (vote)
Assignee: Gentoo Kernel Bug Wranglers and Kernel Maintainers
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2006-08-02 22:43 UTC by lou
Modified: 2006-09-15 11:45 UTC (History)
2 users (show)

See Also:
Package list:
Runtime testing required: ---


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description lou 2006-08-02 22:43:41 UTC
Portage 2.1-r1 (default-linux/x86/no-nptl, gcc-3.4.6, glibc-2.3.6-r4, 2.6.17-gentoo-r4 i686)
=================================================================
System uname: 2.6.17-gentoo-r4 i686 Genuine Intel(R) CPU           T2050  @ 1.60GHz
Gentoo Base System version 1.6.15
app-admin/eselect-compiler: [Not Present]
dev-lang/python:     2.4.3-r1
dev-python/pycrypto: 2.0.1-r5
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2
sys-devel/binutils:  2.16.1-r3
sys-devel/gcc-config: 1.3.13-r3
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-march=pentium-m -pipe -fomit-frame-pointer"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/X11/xkb"
CONFIG_PROTECT_MASK="/etc/env.d /etc/gconf /etc/revdep-rebuild /etc/terminfo"
CXXFLAGS="-O2 -mcpu=i686 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks metadata-transfer sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo"
MAKEOPTS="-j3"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
SYNC="rsync://172.16.0.2/gentoo-portage"
USE="x86 X acpi alsa apache2 apm arts avi berkdb bitmap-fonts bzip2 chroot cli crypt cups curl dba dillo dlloader dri dvd eds emboss encode etheral firefox foomaticdb fortran gd gdbm gif gpm gps gstreamer gtk gtk2 i8x0 imlib innodb isdnlog jpeg libg++ mad motif moznoxft mp3 mpeg ncurses nls ogg opengl opens oss pam pcmcia pcre pdflib perl php png posix pppd pwdb python qt qt3 qt4 quicktime readline reflection samba sdl session snortsam sockets spell spl ssl tcpd truetype truetype-fonts type1-fonts vorbis xface xine xml xmms xorg xv zlib elibc_glibc input_devices_keyboard input_devices_mouse input_devices_evdev kernel_linux userland_GNU video_cards_nvidia"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS, LINGUAS, PORTAGE_RSYNC_EXTRA_OPTS, PORTDIR_OVERLAY


When b44 driver is built as a driver or as a module dmesg shows flow control being off on RX and TX. This results in extremely slow network on the WAN. LAN traffic is normal. Slow traffic never exceeds 15k

Here is dmesg output
b44: eth0: Link is up at 100 Mbps, full duplex.
b44: eth0: Flow control is off for TX and off for RX.

acpi and eth0 do NOT share the same interrupt and adding the following kernel flags do not fix the problem:

acpi=noirq ro acpi=off

I even tried disabling ACPI in the kernel all together. The only way to get b44 to work properly is to roll back to gentoo-sources-2.6.16-r12.
Comment 1 Daniel Drake (RETIRED) gentoo-dev 2006-08-13 03:06:18 UTC
Is this problem reproducible on the latest development kernel, currently 2.6.18-rc4?
Comment 2 lou 2006-08-31 10:01:44 UTC
Hello

I'm experiencing the same behaviour on another Dell box. SC430 with the following nic:
Ethernet controller: Broadcom Corporation NetXtreme BCM5751 Gigabit Ethernet PCI Express (rev 01)

Behaviour started with 2.6.17-r4 and 2.6.17-r7

This box is running vmware-server. My laptop project is on hold, but will try to build 2.6.18-rc4 on it soon.

Here is info on the SC430

Portage 2.1-r2 (default-linux/x86/2005.1, gcc-3.4.6, glibc-2.3.6-r4, 2.6.17-gentoo-r7 i686)
=================================================================
System uname: 2.6.17-gentoo-r7 i686 Intel(R) Pentium(R) 4 CPU 2.80GHz
Gentoo Base System version 1.12.4
app-admin/eselect-compiler: [Not Present]
dev-lang/python:     2.3.5-r2, 2.4.3-r1
dev-python/pycrypto: 2.0.1-r5
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2
sys-devel/binutils:  2.16.1-r3
sys-devel/gcc-config: 1.3.13-r3
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r2
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-march=pentium4 -O2 -pipe"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/X11/xkb"
CONFIG_PROTECT_MASK="/etc/env.d /etc/gconf /etc/revdep-rebuild /etc/terminfo"
CXXFLAGS="-O2 -mcpu=i686 -pipe"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks metadata-transfer sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo"
MAKEOPTS="-j3"
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
PORTDIR_OVERLAY="/usr/local/portage"
SYNC="rsync://172.16.0.2/gentoo-portage"
USE="x86 apm arts avi berkdb bitmap-fonts cli crypt cups dlloader dri eds emboss encode foomaticdb fortran gdbm gif gpm gstreamer imlib ipv6 isdnlog jpeg libg++ libwww mad mikmod motif mp3 mpeg ncurses nls ogg oggvorbis opengl oss pam pcre pdflib perl png pppd python qt3 qt4 quicktime readline reflection sdl session spell spl ssl tcpd truetype truetype-fonts type1-fonts vorbis xml2 xmms xorg xv zlib elibc_glibc input_devices_keyboard input_devices_mouse input_devices_evdev kernel_linux userland_GNU video_cards_apm video_cards_ark video_cards_ati video_cards_chips video_cards_cirrus video_cards_cyrix video_cards_dummy video_cards_fbdev video_cards_glint video_cards_i128 video_cards_i740 video_cards_i810 video_cards_imstt video_cards_mga video_cards_neomagic video_cards_nsc video_cards_nv video_cards_rendition video_cards_s3 video_cards_s3virge video_cards_savage video_cards_siliconmotion video_cards_sis video_cards_sisusb video_cards_tdfx video_cards_tga video_cards_trident video_cards_tseng video_cards_v4l video_cards_vesa video_cards_vga video_cards_via video_cards_vmware video_cards_voodoo"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS, LINGUAS, PORTAGE_RSYNC_EXTRA_OPTS

Linux comp 2.6.17-gentoo-r7 #1 Tue Aug 29 11:39:57 MDT 2006 i686 Intel(R) Pentium(R) 4 CPU 2.80GHz GNU/Linux

MemTotal:      3115216 kB
MemFree:        121840 kB
Buffers:         58348 kB
Cached:        2783208 kB
SwapCached:          0 kB
Active:        1951968 kB
Inactive:       956548 kB
HighTotal:     2226736 kB
HighFree:        34704 kB
LowTotal:       888480 kB
LowFree:         87136 kB
SwapTotal:      979832 kB
SwapFree:       979832 kB
Dirty:           86632 kB
Writeback:           0 kB
Mapped:         510040 kB
Slab:            38124 kB
CommitLimit:   2537440 kB
Committed_AS:   152960 kB
PageTables:       2628 kB
VmallocTotal:   114680 kB
VmallocUsed:      4276 kB
VmallocChunk:   110364 kB


processor       : 0
vendor_id       : GenuineIntel
cpu family      : 15
model           : 4
model name      : Intel(R) Pentium(R) 4 CPU 2.80GHz
stepping        : 9
cpu MHz         : 2793.589
cache size      : 1024 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 5
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe nx lm constant_tsc up pni monitor ds_cpl cid cx16 xtpr lahf_lm
bogomips        : 5594.77




Thanks
Comment 3 lou 2006-09-03 21:06:11 UTC
I was able to fix this on the vmware host by removing tcp_window_scaling

echo 0 > /proc/sys/net/ipv4/tcp_window_scaling

Found this out here:
http://www.uwsg.iu.edu/hypermail/linux/kernel/0606.2/0778.html

Seems to be related with 2.6.17 kernels. I'm not sure if this will address the flowcontrol, but will test shortly.
Comment 4 Daniel Drake (RETIRED) gentoo-dev 2006-09-04 11:31:43 UTC
I'm fairly sure the flow control is unrelated (and shouldn't have changed since earlier kernels). It sounds like there is a broken router somewhere in your line of sight, see http://lwn.net/Articles/92727/

2.6.17 scales the window scaling factor based on your RAM size much more than previous kernels did.
Comment 5 lou 2006-09-04 18:08:43 UTC
Flow control was an issue with the laptop, and this looks to be a separate issue on my vmware-server (I should have broke it out). The change to sysctl with tcp_window_scaling fixed my problem. I have multiple computers, only servers that are having issues are the servers with gentoo-sources-2.6.17-r*. I even have a vmware-client 'on' the vmware-server that is running with hardened-sources-2.6.16 and it works fine through the same interface. That's an interesting article on the routers and window scaling.

Thanks for the reply.
Comment 6 Daniel Drake (RETIRED) gentoo-dev 2006-09-05 06:44:29 UTC
So, are there b44 issues left to solve or has the TCP window scaling thing solved the problem?
Comment 7 Daniel Drake (RETIRED) gentoo-dev 2006-09-08 18:17:26 UTC
Assuming fixed, reopen if not
Comment 8 Tyson Harding 2006-09-15 10:20:22 UTC
I am having the same problem listed here. The tcp_window_scaling does not change how my system reacts. I have updated to the latest development kernel with no luck.

The bcm4400 drivers were working fine, since they have been masked I decided to try the ones in the kernel, but they are not working. I have also tried downloading the drivers from broadcom (1.00g) and they behave the same way (this was expected since they are supposed to be the same drivers now)

The open source kernel drivers do appear to work without a problem on a laptop I have but lspci show it is not an identical controller.

lspci from laptop that does work:
03:00.0 Ethernet controller: Broadcom Corporation BCM4401-B0 100Base-TX (rev 02)

lspci from desktop that does NOT work:
01:09.0 Ethernet controller: Broadcom Corporation BCM4401 100Base-T (rev 01)

As I said, the proprietary bcm4400 driver workes fine on both the desktop and laptop.

I would prefer the bcm4400 package remain unmasked until this problem is resolved. Add blocking of 144314.

Portage 2.1.1 (default-linux/x86/2006.0, gcc-4.1.1, glibc-2.4-r3, 2.6.18-rc7 i686)
=================================================================
System uname: 2.6.18-rc7 i686 Intel(R) Pentium(R) 4 CPU 2.20GHz
Gentoo Base System version 1.12.5
Last Sync: Fri, 15 Sep 2006 11:00:01 +0000
app-admin/eselect-compiler: [Not Present]
dev-java/java-config: 1.2.11-r1
dev-lang/python:     2.3.5-r2, 2.4.3-r1
dev-python/pycrypto: 2.0.1-r5
dev-util/ccache:     [Not Present]
dev-util/confcache:  [Not Present]
sys-apps/sandbox:    1.2.17
sys-devel/autoconf:  2.13, 2.59-r7
sys-devel/automake:  1.4_p6, 1.5, 1.6.3, 1.7.9-r1, 1.8.5-r3, 1.9.6-r2
sys-devel/binutils:  2.16.1-r3
sys-devel/gcc-config: 1.3.13-r3
sys-devel/libtool:   1.5.22
virtual/os-headers:  2.6.11-r5
ACCEPT_KEYWORDS="x86"
AUTOCLEAN="yes"
CBUILD="i686-pc-linux-gnu"
CFLAGS="-march=pentium4 -O2 -pipe -fomit-frame-pointer"
CHOST="i686-pc-linux-gnu"
CONFIG_PROTECT="/etc /usr/share/X11/xkb"
CONFIG_PROTECT_MASK="/etc/env.d /etc/gconf /etc/revdep-rebuild /etc/terminfo"
CXXFLAGS="-march=pentium4 -O2 -pipe -fomit-frame-pointer"
DISTDIR="/usr/portage/distfiles"
FEATURES="autoconfig distlocks metadata-transfer sandbox sfperms strict"
GENTOO_MIRRORS="http://distfiles.gentoo.org http://distro.ibiblio.org/pub/linux/distributions/gentoo"
LINGUAS=""
PKGDIR="/usr/portage/packages"
PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --delete-after --stats --timeout=180 --exclude='/distfiles' --exclude='/local' --exclude='/packages'"
PORTAGE_TMPDIR="/var/tmp"
PORTDIR="/usr/portage"
SYNC="rsync://rsync.gentoo.org/gentoo-portage"
USE="x86 X a52 aac acl alsa apache2 audiofile berkdb bitmap-fonts bl bzip2 cdda cddb cdio cdparanoia cdr cdrom cjk cli crypt css cups divx4linux dlloader dri dts dv dvd dvdread edl eds elibc_glibc emboss encode esd exif expat fam fame fb fbcon ffmpeg fftw flac font-server foomaticdb fortran gd gdbm gif gimp gkrellm glut gmp gnome gpm graphviz gstreamer gtk gtk2 gtkhtml h323 hal id3 idn ieee1394 imlib input_devices_keyboard input_devices_mouse ipod ipv6 isdnlog joystick jpeg jpeg2k kernel_linux lame ldap libg++ libwww live lm_sensors logrotate lzo lzw mad maildir matroska mikmod mjpeg mmx mmxext mng motif mp3 mp4live mpeg mpeg2 mpeg4 mplayer mysql mysqli mythtv nas ncurses network nfs nls nptl nptlonly nvidia ogg openal opengl pam pcre perl player png pppd python qt3 quicktime readline reflection rtc samba sdl session smp sox speex spell spl sse sse2 ssl stream subtitles svg svga tcpd tga theora tiff timidity transcode truetype truetype-fonts type1-fonts udev unicode usb userland_GNU v4l v4l2 vcd vcdimager video_cards_nvidia videos vidix vim vlm vorbis win32codecs wma wma123 wmf x264 xanim xinerama xml xml2 xmms xorg xrandr xv xvid xvmc zlib"
Unset:  CTARGET, EMERGE_DEFAULT_OPTS, INSTALL_MASK, LANG, LC_ALL, LDFLAGS, MAKEOPTS, PORTAGE_RSYNC_EXTRA_OPTS, PORTDIR_OVERLAY
Comment 9 Tyson Harding 2006-09-15 10:22:22 UTC
Oops on my last comment I listed the wrong bug number for blocking. The correct number is 145525.
Comment 10 Daniel Drake (RETIRED) gentoo-dev 2006-09-15 11:45:52 UTC
Please open a new bug for your issue. There is no indication that your problem is any way related to the one reported here (i.e. no indication of TCP window scaling problems, no indicated relation to flow control).