On Jan 26, 2009, at 10:17 PM, Tex wrote:

> I have been running some surveys with the Atheros and Broadcom  
> Drivers.
> When viewing the frames from wireshark , while in monitor mode, I
> noticed that the Atheros driver is reporting FCS errors on some  
> frames,
> but the Broadcom driver isn't, in the radiotap header. If I expand the
> frame and look at the FCS wireshark reports the frame as bad and shows
> what the CRC should be. I put printk's in pio_rx_frame() (see below)  
> and
> noticed that the macstat is being reported as zero. Does this mean  
> that
> the status is not being sent to the driver from the chip or  
> firmware? Is
> there some way for me to configure it to report bad FCS frames?
When the flag B43_MACCTL_KEEP_BAD is enabled in the opmode of b43,  
then the driver activates the corresponding flag in the mac control  
high register so that the firmware is configured to report bad frames  
to the host. However as soon as they are sent through dma, the b43  
driver logs messages like "b43-phy5 debug: RX: Packet dropped" and  
frames are not delivered to the upper layers. To investigate you can  
try to enable that flag by modifying the basic configuration in  
b43_adjust_opmode(struct b43_wldev *dev) defined in main.c.

Cheers,
-FG

>
>             macstat = le32_to_cpu(rxhdr.mac_status);
>             printk (KERN_DEBUG "pio_rx_frame macstat is %x\n",  
> macstat);
>             if (macstat & B43_RX_MAC_FCSERR) {
>                     if (!(q->dev->wl->filter_flags & FIF_FCSFAIL)) {
>                             /* Drop frames with failed FCS. */
>                             err_msg = "Frame FCS error";
>                             goto rx_error;
>                     }
>
> b43_op_configure filter has the following flags set:
>        *fflags &= FIF_PROMISC_IN_BSS |
>                  FIF_ALLMULTI |
>                  FIF_FCSFAIL |
>                  FIF_PLCPFAIL |
>                  FIF_CONTROL |
>                  FIF_OTHER_BSS |
>                  FIF_BCN_PRBRESP_PROMISC;
>
>        changed &= FIF_PROMISC_IN_BSS |
>                   FIF_ALLMULTI |
>                   FIF_FCSFAIL |
>                   FIF_PLCPFAIL |
>                   FIF_CONTROL |
>                   FIF_OTHER_BSS |
>                   FIF_BCN_PRBRESP_PROMISC;
>
>
>
>
> Thanks much
> Tex
>
> _______________________________________________
> Bcm43xx-dev mailing list
> [email protected]
> https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

-------

Francesco Gringoli, PhD - Assistant Professor
Dept. of Electrical Engineering for Automation
University of Brescia
via Branze, 38
25123 Brescia
ITALY

Ph:  ++39.030.3715843
FAX: ++39.030.380014
WWW: http://www.ing.unibs.it/~gringoli




_______________________________________________
Bcm43xx-dev mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

Reply via email to