tcpdump mailing list archives
Re: libpcap capture performance drop
From: Miroslav Lichvar <mlichvar () redhat com>
Date: Mon, 16 Aug 2010 09:59:24 +0200
On Fri, Aug 13, 2010 at 02:00:22PM -0700, Guy Harris wrote:
No. I forgot that the availability of the memory-mapped pcap mechanism isn't determined by the configure script, it's determined by what the system includes offer. Instead, in pcap-linux.c, just #if 0 out the /* check for memory mapped access avaibility. We assume every needed * struct is defined if the macro TPACKET_HDRLEN is defined, because it * uses many ring related structs and macros */
Would it be possible to support disabling mmaped access in runtime with an environment variable? I had one user complaining about slow startup of tcpdump -s 0 and it seems that there is not much we can do about it as kernel has to allocated large blocks of continuous memory. -- Miroslav Lichvar - This is the tcpdump-workers list. Visit https://cod.sandelman.ca/ to unsubscribe.
Current thread:
- libpcap capture performance drop Doktor Bernd (Aug 10)
- Re: libpcap capture performance drop Guy Harris (Aug 13)
- Re: libpcap capture performance drop Doktor Bernd (Aug 15)
- Re: libpcap capture performance drop Guy Harris (Aug 15)
- Re: libpcap capture performance drop Doktor Bernd (Sep 06)
- Re: libpcap capture performance drop Doktor Bernd (Aug 15)
- Re: libpcap capture performance drop Miroslav Lichvar (Aug 16)
- Re: libpcap capture performance drop Guy Harris (Aug 13)