User Tools

Site Tools


r8187

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revisionPrevious revision
Next revision
Previous revision
Next revisionBoth sides next revision
r8187 [2008/07/29 13:47] – Ensure that the rtl8187 module also gets removed, and disregard any error messages given by rmmod. netrolller3dr8187 [2009/04/01 20:51] – enhance SKB BUG fix darkaudax
Line 1: Line 1:
 ====== General ====== ====== General ======
  
-The r8187 driver works properly for the Realtek RTL8187L chipset.  Support for the RTL8187B chipset is under development but is not fully working.+The r8187 driver works properly for the Realtek RTL8187L chipset.  Support for the RTL8187B chipset is under development but is not fully working. See [[r8187b]] for the RTL8187B ieee80211 driver.
  
-This page only deals with the ieee80211 version of the r8187 driver. For the mac80211 rtl8187 version see the [[mac80211|mac80211 page]].  To understand the differences, see [[install_drivers#mac80211_versus_ieee80211_stacks|mac80211 versus ieee80211 stacks]] write-up.+This page only deals with the ieee80211 version of the r8187 driver. For the mac80211 rtl8187 version see the [[rtl8187|rtl8187 page]].  To understand the differences, see [[install_drivers#mac80211_versus_ieee80211_stacks|mac80211 versus ieee80211 stacks]] write-up.
  
 **IMPORTANT**\\ **IMPORTANT**\\
Line 82: Line 82:
  
 ====== Troubleshooting Tips ====== ====== Troubleshooting Tips ======
 +
  
  
Line 88: Line 89:
 If you have the file rtl8187.ko in the /lib/modules directory tree then you have two options to blacklist it.  Failure to do this will mean that the ieee80211 r8187 module described on this page will fail to work properly.  Here are the options: If you have the file rtl8187.ko in the /lib/modules directory tree then you have two options to blacklist it.  Failure to do this will mean that the ieee80211 r8187 module described on this page will fail to work properly.  Here are the options:
  
-  * Move the file to another area on your system as follows then do "depmod -ae" Move /lib/modules/k#/kernel/drivers/net/wireless/mac80211/rtl818x/rtl8187.ko to a safe place.  The "k#" and/or other parts of the path will be different for your distribution/system.  Use "locate 8187.ko" or "find /lib/modules -name *8187*" to find the full path.  After moving it, do "depmod -ae".+  * Move the file to another area on your system as follows then do "depmod -ae" Move /lib/modules/$(uname -r)/kernel/drivers/net/wireless/mac80211/rtl8187.ko to a safe place.  The "$(uname -r)" and/or other parts of the path may be different for your distribution/system.  Use "locate 8187.ko" or "find /lib/modules -name *8187*" to find the full path.  After moving it, do "depmod -ae".
   * Edit /etc/modprobe.d/blacklist and add "blacklist rtl8187" as a new line.   * Edit /etc/modprobe.d/blacklist and add "blacklist rtl8187" as a new line.
  
Line 195: Line 196:
  
 Solution: This typically occurs after you have upgraded your kernel version. Delete the all the patch files and install a fresh version.  You should now be able to compile it successfully. Also ensure that you have matching kernel header files. Solution: This typically occurs after you have upgraded your kernel version. Delete the all the patch files and install a fresh version.  You should now be able to compile it successfully. Also ensure that you have matching kernel header files.
 +
  
  
Line 206: Line 208:
   * Having legacy USB support enabled in your BIOS. Try disabling this option.   * Having legacy USB support enabled in your BIOS. Try disabling this option.
  
-**Note:** VMWare Workstation 6.0 and VMWare Player 2.0 use USB 1.1 and when injecting at high speed, injection uses all USB bandwith (so, nearly no packets are received) and you can notice a lot of packets lost in airodump-ng.+**Note:** VMWare Workstation 6.0 and VMWare Player 2.0 use USB 1.1 and when injecting at high speed, injection uses all USB bandwith (so, nearly no packets are received) and you can notice a lot of packets lost in airodump-ng.
  
  
Line 238: Line 240:
 ===== Optimizing injection rates ===== ===== Optimizing injection rates =====
 Some people have reported that including "-x 250" on the aireplay-ng command optimizes their injection rates.  You will have to experiment to see if this helps you or not. Some people have reported that including "-x 250" on the aireplay-ng command optimizes their injection rates.  You will have to experiment to see if this helps you or not.
 +
  
  
Line 248: Line 251:
 The root cause of these messages is not known at this point in time.  They can be hidden by the following means: The root cause of these messages is not known at this point in time.  They can be hidden by the following means:
  
-  * To remove them from the console: "dmesg -2"+  * To remove them from the console: "dmesg -n2"
   * To remove them from system log files, edit /etc/syslog.conf or /etc/rsyslog.conf depending on your system.  Add "kern.!debug" to definition of each file receiving the debug messages.   * To remove them from system log files, edit /etc/syslog.conf or /etc/rsyslog.conf depending on your system.  Add "kern.!debug" to definition of each file receiving the debug messages.
 +
 +Another more radical approach is to comment out the kernel messages in the kernel source code.  This will also mean recompiling your kernel.
 +
 +Change net/core/skbuff.c in your kernel source:
 +
 +   void skb_truesize_bug(struct sk_buff *skb)
 +   {
 +         printk(KERN_ERR "SKB BUG: Invalid truesize (%u) "
 +                "len=%u, sizeof(sk_buff)=%Zd\n",
 +                skb->truesize, skb->len, sizeof(struct sk_buff));
 +   }
 +   EXPORT_SYMBOL(skb_truesize_bug);
 +
 +to:
 +
 +   void skb_truesize_bug(struct sk_buff *skb)
 +   {
 +   (void)skb;
 +   //      printk(KERN_ERR "SKB BUG: Invalid truesize (%u) "
 +   //             "len=%u, sizeof(sk_buff)=%Zd\n",
 +   //             skb->truesize, skb->len, sizeof(struct sk_buff));
 +   }
 +   EXPORT_SYMBOL(skb_truesize_bug);
 +
 +This effectively eliminates the kernel from reporting the SKB BUG messages.
 +
 +===== "iwe_stream_add_event" compile error message =====
 +
 +If you get a series of compile messages similar to "error: passing argument 1 of 'iwe_stream_add_event' from incompatible pointer type" then do the following:
 +
 +Use
 +
 +  wget http://patches.aircrack-ng.org/rtl8187_2.6.27.patch
 +
 +instead of
 +
 +  wget http://patches.aircrack-ng.org/rtl8187_2.6.24v3.patch
 +
 +in the instructions at the top of the page.
 +
 +===== "asm/semaphore.h: No such file or directory" compile error message =====
 +
 +See this [[http://tinyshell.be/aircrackng/forum/index.php?topic=4305.msg24483#msg24483|forum entry]]:
 +
 +Please note a repair apparently needed in ' rtl8187_2.6.27.patch' for kernel 2.6.27 +/-
 +
 +Unmodified, you will get the following error :
 + -- output from ' make ' :
 +  ...In file included from /usr/src/drivers/rtl8187_linux_26.1010.0622.2006/beta-8187/r8187_core.c:65:
 +     /usr/src/drivers/rtl8187_linux_26.1010.0622.2006/beta-8187/r8187.h:47:27: error: asm/semaphore.h: No such file or directory
 +
 +
 +Modification to file    r8187.h   :
 +    lines 46,47 are :
 +      #include <asm/io.h>
 +      #include <asm/semaphore.h>
 +
 +    overwrite lines 46,47 to this....
 +      #if (LINUX_VERSION_CODE < KERNEL_VERSION(2,6,19))
 +        #include <asm/io.h>
 +        #include <asm/semaphore.h>
 +      #else
 +        #include <linux/io.h>
 +        #include <linux/semaphore.h>
 +      #endif
  
  
r8187.txt · Last modified: 2017/01/09 21:35 by mister_x