Porteus 32 bit is worse then 64 bit?

[komoornik]

Device: Asus 1215n

Broadcom wifi card

Clean boot of 32 bit Porteus without any modifications - wicd can't find any networks

Clean boot of 64 bit version - wicd finds and connects to networks

Question: Why the 32 bit version is worse?

[fanthom]

seems that broadcom-sta driver was not recompiled against 2.6.38.8 kernel in 32bits.
i have added updated version to 32bit fixes:
http://ponce.cc/porteus/i486/testing/1. ... .0-fix.xzm
copy it to your /porteus/modules folder and reboot system - should be ok then

[komoornik]

Not working

I've tried: ifconfig wlan0 up, and it says no such device or smth like that.

[Hamza]

Did you tried to load the module before to try to use your wireless device ?

[komoornik]

What do you mean? I thought it's loaded automatically when if i put it in modules folder and reboot?

[Hamza]

When you add a new module (.ko file) to the kernel, You must load it otherwise it will not be used.

[fanthom]

@komoornik
rebooting should do the job but you can load driver manually as advised by Hamza. open terminal and type 'modprobe wl' then run 'tail -n50 /var/log/dmesg' and past the output here. to be sure that wl driver was loaded correctly run 'lsmod | grep wl' and show us the result.

[komoornik]

Code: Select all

bash-4.1# tail -n50 /var/log/dmesg
[   12.162741] [drm] Initialized drm 1.1.0 20060810
[   12.177154] brcm80211 0000:02:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
[   12.177170] brcm80211 0000:02:00.0: setting latency timer to 64
[   12.211405] brcm80211: fail to load firmware brcm/bcm43xx-0.fw
[   12.211414] brcm80211: Failed to find firmware usually in /lib/firmware/brcm
[   12.211463] brcm80211 0000:02:00.0: PCI INT A disabled
[   12.211494] brcm80211: wl_pci_probe: wl_attach failed!
[   12.241420] lib80211: common routines for IEEE802.11 drivers
[   12.241428] lib80211_crypt: registered algorithm 'NULL'
[   12.393713] i915 0000:00:02.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[   12.393724] i915 0000:00:02.0: setting latency timer to 64
[   12.444240] mtrr: type mismatch for b0000000,10000000 old: write-back new: write-combining
[   12.444249] [drm] MTRR allocation failed.  Graphics performance may suffer.
[   12.445366] i915 0000:00:02.0: irq 44 for MSI/MSI-X
[   12.445378] [drm] Supports vblank timestamp caching Rev 1 (10.10.2010).
[   12.445384] [drm] Driver supports precise vblank timestamp query.
[   12.452877] [drm:intel_dsm_platform_mux_info] *ERROR* MUX INFO call failed
[   12.453032] [drm:intel_dsm_platform_mux_info] *ERROR* MUX INFO call failed
[   12.526719] wl: module license 'unspecified' taints kernel.
[   12.526726] Disabling lock debugging due to kernel taint
[   12.534743] vgaarb: device changed decodes: PCI:0000:00:02.0,olddecodes=io+mem,decodes=none:owns=io+mem
[   12.534753] vgaarb: transferring owner from PCI:0000:00:02.0 to PCI:0000:04:00.0
[   12.535226] [drm] initialized overlay support
[   12.539295] wl 0000:02:00.0: PCI INT A -> GSI 17 (level, low) -> IRQ 17
[   12.539314] wl 0000:02:00.0: setting latency timer to 64
[   12.575396] lib80211_crypt: registered algorithm 'TKIP'
[   12.575730] eth0: Broadcom BCM4727 802.11 Hybrid Wireless Controller 5.100.82.38
[   12.602170] udev[2029]: renamed network interface eth0 to eth2
[   12.793557] checking generic (b0000000 7f0000) vs hw (b0000000 10000000)
[   12.793567] fb: conflicting fb hw usage inteldrmfb vs VESA VGA - removing generic driver
[   12.793633] Console: switching to colour dummy device 80x25
[   12.796146] fbcon: inteldrmfb (fb0) is primary device
[   12.800109] Console: switching to colour frame buffer device 170x48
[   12.847224] fb0: inteldrmfb frame buffer device
[   12.847231] drm: registered panic notifier
[   12.847367] [Firmware Bug]: ACPI(GFX0) defines _DOD but not _DOS
[   12.848850] input: Video Bus as /devices/LNXSYSTM:00/device:00/PNP0A08:00/device:26/LNXVIDEO:00/input/input10
[   12.849280] ACPI: Video Device [GFX0] (multi-head: yes  rom: yes  post: no)
[   12.859923] ACPI Warning: _BQC returned an invalid level (20110112/video-473)
[   12.869966] acpi device:2c: registered as cooling_device4
[   12.871473] input: Video Bus as /devices/LNXSYSTM:00/device:00/PNP0A08:00/LNXVIDEO:01/input/input11
[   12.871879] ACPI: Video Device [VGA] (multi-head: yes  rom: no  post: no)
[   12.872991] [drm] Initialized i915 1.6.0 20080730 for 0000:00:02.0 on minor 0
[   12.873150] HDA Intel 0000:00:1b.0: PCI INT A -> GSI 22 (level, low) -> IRQ 22
[   12.873282] HDA Intel 0000:00:1b.0: irq 45 for MSI/MSI-X
[   12.873336] HDA Intel 0000:00:1b.0: setting latency timer to 64
[   12.924243] hda_codec: ALC269VB: BIOS auto-probing.
[   12.927463] input: HDA Intel Mic as /devices/pci0000:00/0000:00:1b.0/sound/card0/input12
[   12.927762] input: HDA Intel Headphone as /devices/pci0000:00/0000:00:1b.0/sound/card0/input13
[   14.024777] FAT: utf8 is not a recommended IO charset for FAT filesystems, filesystem will be case sensitive!
bash-4.1# lsmod | grep wl
wl                   2538388  0
lib80211                2076  2 lib80211_crypt_tkip,wl

Not working anyway.

[fanthom]

@komornik
for some reason your wifi device gets named as eth0 and not wlan0:

Code: Select all

[   12.575730] eth0: Broadcom BCM4727 802.11 Hybrid Wireless Controller 5.100.82.38

i think wicd could be confused with this name as probably it "thinks" that it's a wired NIC.

you can use 'pns-tool' to set your network with eth0 or try to fight with wicd.
in second case pls add 'brcm80211' 'b43' and 'b43legacy' drivers to /etc/modprobe.d/blacklist.conf to make sure that they are not loaded during boot.
let me know if any of my advices worked for you.

[francois]

Broadcom are really dirty drivers to run on linux. Since there is linux.sta, everything should work better though.

From an old thread for remix (googling broadcom francois.e remix keywords). I have a broadcom 4328 on my hp pavilion 2713ca:
http://slax.speedymirror.com/forum.php? ... ntID=70793

Citation:
"Good, Wifi is finally working for me. I knew that linux sta for bcm was on the distribution. I finally understood that to use the bcm43xx (for me 4328), all I had to do was to blaclist ssb and b43, as proposed in:
https://wiki.archlinux.org/index.php/Broadcom_BCM43XX"

Hoping that this helps.

[brokenman]

If you choose to go into battle with wicd here is some ammunition.

You may want to specify which MAC gets which device name using udev and a rule such as:

KERNEL=="eth*", SYSFS{address}=="00:13:32:fo:ob:ar", NAME="eth0"

[komoornik]

I think i can change wlan0 to eth0 in wicd.

Anyway, my question would be why it works nicely on 64 bit version ?

[fanthom]

my question would be why it works nicely on 64 bit version ?
i can see 2 possibilities:
1) if you are saving changes then 'wlan0' name could be linked with previous driver - try 'Always Fresh' mode.
2) i have seen strange scenario when 'modesetting' DDX driver was taking priority over 'nouveau' when both were placed in 002-xorg module. once nouveau was moved to 003 or later it was loaded first by Xorg. i can't explain why it was like that.
in your case i would try to remaster 000-kernel.xzm and replace faulty wl.ko driver with good one from the fix.
if you have some free time you could try this experiment.

[francois]

What about blacklisting ssb and b43?
However, it seems I have some problems too :
http://porteus.org/forum/viewtopic.php?f=46&t=758

[francois]

On the topic of 32-bit worse than 64-bit, I am not so sure. It might depend on which features, I had less issues on Porteus 32-bit v. 0.9. See What is your computer model:
http://forum.porteus.org/faq.php?sid=4a ... 312a39c3f0