Change Details

We can test a lot of the sack_filter changes in user space (which I have yet to do so there may be updates coming to this). Then I will also develop some packet drill scripts which I will insert here which can be used in conjunction with BBlogs that things are working as we expect i.e. normal sacks work properly and illicit ones are ignored.

We can test a lot of the sack_filter changes in user space (which I have yet to do so there may be updates coming to this). Then I will also develop some packet drill scripts which I will insert here which can be used in conjunction with BBlogs that things are working as we expect i.e. normal sacks work properly and illicit ones are ignored. Ok I have validated now by compiling sack_fliter in user space and playing with all the corner cases and the general. It nicely blocks single byte sacks of most flavors unless we are compressing the map. I have refined the end tests too so you can't attack at the end using 1 byte at a time that spans to the sndmax Still have yet to build the pkt_drill script that will be here to test the low mss case i.e. to validate we just turn sack off. Will do that early next week.

We can test a lot of the sack_filter changes in user space (which I have yet to do so there may be updates coming to this). Then I will also develop some packet drill scripts which I will insert here which can be used in conjunction with BBlogs that things are working as we expect i.e. normal sacks work properly and illicit ones are ignored. Ok I have validated now by compiling sack_fliter in user space and playing with all the corner cases and the general. It nicely blocks single byte sacks of most flavors unless we are compressing the map. I have refined the end tests too so you can't attack at the end using 1 byte at a time that spans to the sndmax Still have yet to build the pkt_drill script that will be here to test the low mss case i.e. to validate we just turn sack off. Will do that early next week.