[ANN]: Microolap Packet Sniffer SDK 4.0

Greetings,

The major release 4.0 of Packet Sniffer SDK is out: Ethernet emulation mode for Loopback traffic, Adjustable packets pool, support of x64(AMD64, EM64T) systems in BPF JIT compiler, packets marking in BPF filter, the queue for packets sending.....

Also WinPCap to PSSDK migration module is significantly improved: http://microolap.com/products/network/pssdk/helpdll/WinPcap/WinPcapMigration.htm

Product page: http://microolap.com/products/network/pssdk/

Docs online:
VCL - http://microolap.com/products/network/pssdk/helpvcl/
DLL/static libraries - http://microolap.com/products/network/pssdk/helpdll/

Docs offline (CHM):
VCL - http://microolap.com/downloads/pssdk/help/pssdkvcl_en.zip
DLL/static libraries - http://microolap.com/downloads/pssdk/help/pssdkdll_en.zip

Coupon code for 20% discount - "Christmas" (without quotes, of course).

Merry Christmas and happy New Year!

---
Best regards,
MicroOLAP Technologies LTD
http://www.microolap.com

======================================
Full changelog:

Version 4.0, December 11, 2007.

* Now BPF JIT compiler supports x64(AMD64, EM64T) systems, that significantly accelerates traffic processing on 64-bit machines.

* BPF instructions set is extended: BPF_HTONL, BPF_HTONS, BPF_MARK instructions are added (see BPF Instructions). Now you can create BPF filters which will mark each packet with desired value associated with it, that significantly simplifies further processing of filtered traffic (see also HNPacket.Mark[Get/Set] property).

* Ethernet emulation mode for HNLBAdapter object is added. Now Loopback adapter allows to process both local and remote TCP/UDP traffic the same way as it is possible for traffic captured from a physical network device. All objects, properties, methods, and events available for HNAdapter object now are available for Loopback adapter too.

* The queue for packets sending for HNAdapter object is added (see HNAdapter.SendQueue[Get/Set]). This feature is the best alternative for AdpSyncSend and AdpSyncSendEx functions, and allows to arrange high-performance packets sending in very flexible and powerful way.

* Adjustable packets pool for HNAdapter and HNLBAdapter is added. Now, changing HNAdapter.PacketPool[Get/Set] or HNLBAdapter.PacketPool[Get/Set] properties you can optimize both your application performance and the memory required for it.

* LostCount statistic parameter indicating the quantity of packets lost by PSSDK internal drivers for HNAdapter and HNLBAdapter objects is added (see AdpGetLostCount and LbAdpGetLostCount functions). This parameter will help you to detect performance bottlenecks in your application.

* TimeStamping has been standardized. Now under Windows NT/2000/XP/2003/Vista both in standard mode and HighPrecisionTime mode packets are timestamped with the value equal to hundreds of nanoseconds since January 1, 1601.

* HNAdapterConfig.IsWireless[Get] property for HNAdapterConfig object is added. This property indicates that the network adapter actually is a wireless (802.11) device.

* Real-time packets sorting by ID in the HNQueue object significantly improved.

* BPF instructions parser implementation in the HNUserFilter object optimized.

* Winpcap to PSSDK migration module is improved and expanded. Now WinPcap-based applications recompiled with WinPcap Wrapper can capture Loopback traffic.


.

Relevant Pages

  • Re: bge dropping packets issue
    ... listening in this instance is using BPF. ... to understand how much can BPF take before dumping packets due to lack ... bpf is turned on on the receiver. ... packets at a send rate of about 400 kpps. ...
    (freebsd-net)
  • Re: bpf/pcap are weird
    ... BIOCIMMEDIATE mode means "make incoming packets readable immediately; ... might want to try to get as many packets as possible per read on the BPF ... BIOCIMMEDIATE mode would probably still want to block, rather than spin, ... store buffer fills *OR* the timeout expires - then it does need to be ...
    (freebsd-arch)
  • Re: memory mapped packet capturing - bpf replacement ?
    ... > the point at which packets started to be dropped. ... A bit offtopic - what traffic generator you use? ... > still got the same number of copies as vanilla BPF. ... I am copying whole mbuf chain the same way BPF does. ...
    (freebsd-hackers)
  • bpf writes on tun device
    ... I would like to use the BPF device to write packets on a tun device, ... 5.X) says that BPF writes are only supported on Ethernet and SLIP links. ... I can successfully write BPF packets up to 1500 bytes in size (1496 IP ...
    (freebsd-net)