headers
王树桥 | 1 Dec 2004 01:36

The DsrRouteTable document

Hi everyone:

    Why can't I find the documentation of DsrRouteTable in the Click documentation
(http://www.pdos.lcs.mit.edu/click/doc/)? And how can I find it?

    Thank you!
Permalink | Reply | 2 comments |
headers
John Bicket | 1 Dec 2004 04:32
Picon

Re: The DsrRouteTable document

It's probably because the documentation wasn't built with --enable-wifi.
Uou can just look at the file elements/grid/dsrroutetable.hh
in a cvs checkout.
(or at
http://cvs.pdos.lcs.mit.edu/cvs/click/release/one/elements/grid/dsrroutetable.hh)

--john

?????? [wangshuqiao <at> 263.sina.com] wrote:
>Hi everyone:
>
>    Why can't I find the documentation of DsrRouteTable in the Click documentation
(http://www.pdos.lcs.mit.edu/click/doc/)? And how can I find it?
>
>    Thank you!
>
>
>
>_______________________________________________
>click mailing list
>click <at> amsterdam.lcs.mit.edu
>https://amsterdam.lcs.mit.edu/mailman/listinfo/click
Permalink | Reply | 1 comment |
headers
Kishore Ramachandran | 1 Dec 2004 06:43
Picon
Favicon

Question regarding SetTXPower element usage.

Hi All:

I would like to use this element but I cannot find documentation for the
same. Currently I am using it as follows:

SetTXPower(POWER a) where a is a constant between 0 and 63 (I looked at
the source code for this element and saw this range there).

However, when I observe the RSSI using a signal analyzer, there seems to
be no noticeable change. Am I using this correctly? Is there any
documentation for the same? What does this range correspond to?

My script (the one generated by madwifi-stripped/gen_config_ap.click with
the addition of this one line for SetTXPower) installs without complaints
and functions smoothly as well.

Thanks,
Kishore

On Mon, 29 Nov 2004, Beyers Cronje wrote:

> Hi all,
>
> I get an intermittend "Unable to handle kernel NULL pointer
> dereference" when using click-uninstall.
>
> Log file and running click config shown at the bottom.
> I'd say around 90% of the time click-uninstall works 100%.
>
> I'm running one custom element named "ipStats". Ive
(Continue reading)

Permalink | Reply | 1 comment |
headers
John Bicket | 1 Dec 2004 17:34
Picon

Re: Question regarding SetTXPower element usage.

Hi,

Unfortunately there is no support for transmit power control in the
atheros hal currently, and it just disregards the power value.

--john

Kishore Ramachandran [ramack <at> eden.rutgers.edu] wrote:
>Hi All:
>
>I would like to use this element but I cannot find documentation for the
>same. Currently I am using it as follows:
>
>SetTXPower(POWER a) where a is a constant between 0 and 63 (I looked at
>the source code for this element and saw this range there).
>
>However, when I observe the RSSI using a signal analyzer, there seems to
>be no noticeable change. Am I using this correctly? Is there any
>documentation for the same? What does this range correspond to?
>
>My script (the one generated by madwifi-stripped/gen_config_ap.click with
>the addition of this one line for SetTXPower) installs without complaints
>and functions smoothly as well.
>
>Thanks,
>Kishore
>
>On Mon, 29 Nov 2004, Beyers Cronje wrote:
>
>> Hi all,
(Continue reading)

Permalink | Reply | 0 comments |
headers
Eddie Kohler | 1 Dec 2004 20:39
Favicon

Re: The DsrRouteTable document

John Bicket wrote:
> It's probably because the documentation wasn't built with --enable-wifi.

Oops!  Well it is now, thanks for the note.

Hey John, got a couple free hours? :)  Looks like some of the documentation 
uses unexpected formatting.

http://www.pdos.lcs.mit.edu/click/doc/AutoRateFallback.n.html

And every Wifi -> Ethernet element has the same summary ("Turns 80211 packets 
into ethernet packets encapsulates packets in Ethernet header").

Eddie

> Uou can just look at the file elements/grid/dsrroutetable.hh
> in a cvs checkout.
> (or at
> http://cvs.pdos.lcs.mit.edu/cvs/click/release/one/elements/grid/dsrroutetable.hh)
> 
> --john
> 
> 
> ?????? [wangshuqiao <at> 263.sina.com] wrote:
> 
>>Hi everyone:
>>
>>   Why can't I find the documentation of DsrRouteTable in the Click documentation
(http://www.pdos.lcs.mit.edu/click/doc/)? And how can I find it?
>>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Nick Mehor | 2 Dec 2004 07:33
Picon
Favicon

vmalloc in click

Hello Eddie,

  i am trying to use vmalloc from within one of my
elements to allocate a large block of contigous
memory.  Unfortunately, it is failing for me with an
unresolved symbol error.

/usr/local/lib/click.o: unresolved symbol
_Z9__vmallocmi8pgprot_t
click-install: `/sbin/insmod /usr/local/lib/click.o'
failed

>From reading the mailing list it seems like others
have
successfully used vmalloc from within click.  Any
ideas
what might be wrong?

gcc version is 3.2 and kernel is linux 2-4.18.

Thanks in advance
-Nick

		
__________________________________ 
Do you Yahoo!? 
Meet the all-new My Yahoo! - Try it today! 
http://my.yahoo.com
Permalink | Reply | 1 comment |
headers
Eddie Kohler | 2 Dec 2004 09:22
Favicon

Re: vmalloc in click

Hi Nick,

Sounds like you didn't #include <linux/vmalloc.h>; or if you did, that 
you didn't enclose it in click's

#include <click/cxxprotect.h>
CLICK_CXX_PROTECT
....
CLICK_CXX_UNPROTECT
#include <click/cxxunprotect.h>

Because it looks like the vmalloc() symbol is getting C++ linkage (which 
is incorrect).

Eddie

Nick Mehor wrote:
> Hello Eddie,
> 
>   i am trying to use vmalloc from within one of my
> elements to allocate a large block of contigous
> memory.  Unfortunately, it is failing for me with an
> unresolved symbol error.
> 
> /usr/local/lib/click.o: unresolved symbol
> _Z9__vmallocmi8pgprot_t
> click-install: `/sbin/insmod /usr/local/lib/click.o'
> failed
> 
>>From reading the mailing list it seems like others
(Continue reading)

Permalink | Reply | 0 comments |
headers
John Bicket | 2 Dec 2004 21:08
Picon

linuxmodule Packet::make fix

I just checked in a fix for a bug that showed up when using FromHost in
the kernel.   If you've ever seen those messages that say "wmem_alloc !=
0", it was caused by Packet::make not calling the sk_buff destructor
(ie skb_orphan).

As a result, the packet would get recycled in click, and then the socket
would not be able to keep track of the oustanding sk_buff's it allocated
that are still somewhere in the kernel.

--john

----- Forwarded message from John Bicket <jbicket <at> amsterdam.lcs.mit.edu> -----

Subject: [Click-commits] release/one/include/click
From: John Bicket <jbicket <at> amsterdam.lcs.mit.edu>
To: click-commits <at> amsterdam.lcs.mit.edu
Date: Thu, 2 Dec 2004 15:03:13 -0500 (EST)

Update of /home/am0/click/cvsroot/release/one/include/click
In directory amsterdam.lcs.mit.edu:/tmp/cvs-serv84807

Modified Files:
	packet.hh 
Log Message:
orphan packets when we get them from linux; this ensures that sockets can 
do their bookkeeping correctly.

_______________________________________________
Click-commits mailing list
Click-commits <at> amsterdam.lcs.mit.edu
(Continue reading)

Permalink | Reply | 0 comments |
headers
Bart Braem | 3 Dec 2004 16:22
Picon

Re: Perhaps trivial question

Hello Eddie,

I merged my (ns)click version of erase with the cvs version and it works now. 
Thanks!

Bart

Op zondag 28 november 2004 20:43, schreef u:
> Hi Bart,
>
> I think you might have an old version of Click.  The current
> Vector<T*>::erase() has an explicit cast, which should be enough.  Can you
> update to the current anonymous CVS and let us know if that helps?
>
> Eddie
>
> Bart Braem wrote:
> > Hello,
> >
> > In my code I need to traverse a whole Hashmap, use the data inside each
> > entry and then delete that entry. I tried using the erase method based on
> > 1 iterator (later also with 2 iterator arguments) but it doesn't compile:
> > my iterator is incompatible with void**:
> > ../include/click/vector.hh: In member function `T**
> >    Click::Vector<T*>::erase(T**) [with T = Click::Packet]':
> > ../elements/aodv/aodv_waitingfordiscovery.cc:156:   instantiated from
> > here ../include/click/vector.hh:239: error: invalid conversion from
> > `void**' to ` Click::Packet**'
> >
> >
(Continue reading)

Permalink | Reply | 0 comments |
headers
Zhiwei Yu | 4 Dec 2004 06:29
Picon
Favicon

token bucket and leaky bucket implementation

Hi, 

I am click newbie looking for a token bucket and leaky
bucket solution for regulator.  Is there is a way to
get eithor token or leaky solution by using existing
elements?  I found a Queue followed by a Meter can
probably simulate a leaky bucket.  But so far, I
cannot figure out how to deploy a token bucket.  Maybe
I need to implement my own element?  Thank you in
advance.

Regards,
Zhiwei

__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com
Permalink | Reply | 0 comments |

Gmane