site_archiver@lists.apple.com Delivered-To: darwin-dev@lists.apple.com Did you check to see if there was another mbuf in the chain (mbuf_next)? -josh On Jun 5, 2008, at 12:58 AM, Thorsten Kramp wrote: Hello all, Best, Thorsten _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/jgraessley%40apple.com _______________________________________________ Do not post admin requests to the list. They will be ignored. Darwin-dev mailing list (Darwin-dev@lists.apple.com) Help/Unsubscribe/Update your Subscription: http://lists.apple.com/mailman/options/darwin-dev/site_archiver%40lists.appl... I'm developing a TCP filter NKE for 10.5, with a user-level process communicating with the NKE via a control socket in stream mode. The user-level process waits for data from the NKE by calling select and read/write, respectively. However, sometimes when the user-level process writes to the control socket some data, not all the data seems to be passed to the NKE (even though the return value of write indicates that all bytes have been passed to the socket). More precisely, for some reason the NKE only gets the first 204 bytes in the m parameter of the NKE's ctl_send_func. Interestingly, it's always precisely 204 bytes even though the user-level process sometimes writes 217 bytes, 218 bytes, 300 bytes... Does anybody has ever seen this behaviour? All suggestions are very welcome! Thanks in advance. This email sent to jgraessley@apple.com This email sent to site_archiver@lists.apple.com