another attempt to enable memory mapped access to ethernet devicefor linux

["Paolo Abeni" <paolo.abeni () telecomitalia it>]

hello,

following the discussion on the implementation of the memory mapped
access for linux, I rework my previous patch to avoid abusing the
current API. It now try to use a default ring size of 4M, reducing it's
size if the allocation of the ring fails.

I did not add any new pcap function to set the ring size, because I
realized that this feature could be useful even for platforms using bpf,
but the use of a separate function to set the ring size could rise some
portability issues (with bpf the buffer size can't be changed after the
socket is bound to an interface). 

I think that a good solution could be to add a new function to open live
capture with 'extended' options like the capture ring size (as suggested
earlier by Gianluca), while the 'standard' pcap_open_live will use some
reasonable default.
Again, I appreciate any comments/suggestions on said subject.

ciao,

Paolo


--------------------------------------------------------------------

CONFIDENTIALITY NOTICE

This message and its attachments are addressed solely to the persons above and may contain confidential information. If 
you have received the message in error, be informed that any use of the content hereof is prohibited. Please return it 
immediately to the sender and delete the message. Should you have any questions, please contact us by replying to 
webmaster () telecomitalia it.

        Thank you

                                        www.telecomitalia.it

--------------------------------------------------------------------

Attachment: linux_enable_mmap.patch
Description:

-
This is the tcpdump-workers list.
Visit https://cod.sandelman.ca/ to unsubscribe.