Comments on: Fast kernel debugging for VMware, part 5: Bridging the Gap to DbgEng.dll http://www.nynaeve.net/?p=172 Adventures in Windows debugging and reverse engineering. Mon, 23 Jul 2012 05:58:29 +0000 hourly 1 http://wordpress.org/?v=3.0.4 By: Skywing http://www.nynaeve.net/?p=172&cpage=1#comment-41266 Skywing Sat, 25 Jul 2009 15:45:11 +0000 http://www.nynaeve.net/?p=172#comment-41266 Koby: You can use kdsrv to remote kd over the network. (You still need a second box to make the actual physical connection to the debugged machine, however.) Be warned that performance is often poor over WAN links as this remotes the whole KD protocol (which is one packet at a time and needing retransits, ACKs) as far as I know. Koby: You can use kdsrv to remote kd over the network. (You still need a second box to make the actual physical connection to the debugged machine, however.)

Be warned that performance is often poor over WAN links as this remotes the whole KD protocol (which is one packet at a time and needing retransits, ACKs) as far as I know.

]]> By: Koby Kahane http://www.nynaeve.net/?p=172&cpage=1#comment-37464 Koby Kahane Fri, 12 Jun 2009 05:47:30 +0000 http://www.nynaeve.net/?p=172#comment-37464 Does kd actually work as a KDCOM client over TCP in the latest DTW? Consider the scenario presented in this post: http://blog.neocleus.com/2009/06/11/debug-xen-hosted-windows-kernel-over-network/ If kd can do KDCOM over TCP directly, there would be no need for the tcp2pipe proxy they mention. However, my attempts to invoke kd -k with a TCP connection string resulted in invalid parameter errors. Does kd actually work as a KDCOM client over TCP in the latest DTW?

Consider the scenario presented in this post:
http://blog.neocleus.com/2009/06/11/debug-xen-hosted-windows-kernel-over-network/

If kd can do KDCOM over TCP directly, there would be no need for the tcp2pipe proxy they mention. However, my attempts to invoke kd -k with a TCP connection string resulted in invalid parameter errors.

]]> By: Nynaeve » Blog Archive » Viridian guest hypercall interface published http://www.nynaeve.net/?p=172&cpage=1#comment-19908 Nynaeve » Blog Archive » Viridian guest hypercall interface published Sat, 10 Nov 2007 04:03:56 +0000 http://www.nynaeve.net/?p=172#comment-19908 [...] debugging hypercalls, as currently defined, are in fact very much reminiscent of how I originally implemented VMKD. The hypercalls define an interface for a partition to send large chunks of kernel debugger data [...] [...] debugging hypercalls, as currently defined, are in fact very much reminiscent of how I originally implemented VMKD. The hypercalls define an interface for a partition to send large chunks of kernel debugger data [...]

]]> By: Nynaeve » Blog Archive » Fast kernel debugging for VMware, part 6: Roadmap to Future Improvements http://www.nynaeve.net/?p=172&cpage=1#comment-18844 Nynaeve » Blog Archive » Fast kernel debugging for VMware, part 6: Roadmap to Future Improvements Thu, 11 Oct 2007 12:00:22 +0000 http://www.nynaeve.net/?p=172#comment-18844 [...] Nynaeve Adventures in Windows debugging and reverse engineering. « Fast kernel debugging for VMware, part 5: Bridging the Gap to DbgEng.dll [...] [...] Nynaeve Adventures in Windows debugging and reverse engineering. « Fast kernel debugging for VMware, part 5: Bridging the Gap to DbgEng.dll [...]

]]>