Talk:Hammerhead/Protocol

From Openmoko

Revision as of 14:44, 8 February 2009 by Bibo (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Um... Pavel, maybe I'm missing something here, but I think what you're calling "received packet 00" is what I'm calling packet 10. Are you saying that the high nibble of the packet type doesn't matter? Because I can't see any particular connection between these packets and the type 00 packets that gltt sends. Likewise what you're calling packet 03 has a type byte of 23, and 0D has a type byte of 9D. (I'm pretty sure that the high nibble is important. For another example, look at packet type 0E versus 9E.) FM 23:24, 6 May 2007 (CEST)

Since you never answered, I fixed the headings. I don't have much time to work on this at the moment, so that's basically all I did. Also I'm not sure what you're talking about w.r.t packet 06 (/16?) so I left that alone. If you disagree with my analysis, that's fine, and please say so. FM 13:32, 17 May 2007 (CEST)

I also think the high nibble of the packet type is important. 0E seems to be an answer to the 00 packet when a satellite is locked, whereas 9E is one of the possible answers (9D,9E,9F) to the 05 packet. I have started to analyze the cold start trace at [1], there are some statistics about the 00 packet at [2] and a closer look on the sequence of events on a single channel (ch 1) [3]. I used a shell script [4] to filter the serial communication out of the strace output. PH5 16:19, 17 May 2007 (CEST)

Except for the flags, the packet format is now mostly understood [5]. The 'packet type' is really an initial register offset, every channel has its own set of registers, and the 0x2 flag seems to switch to a completely different register bank. PH5 20:58, 22 June 2007 (CEST)

Is there need for a SUPL-Client

Hi folks, i have started a project at jsupl.sourceforge.net to make a free implementation of the SUPL protocol. Is there any need at the openmoko project for this? Regards, Lukas

Personal tools

Um... Pavel, maybe I'm missing something here, but I think what you're calling "received packet 00" is what I'm calling packet 10. Are you saying that the high nibble of the packet type doesn't matter? Because I can't see any particular connection between these packets and the type 00 packets that gltt sends. Likewise what you're calling packet 03 has a type byte of 23, and 0D has a type byte of 9D. (I'm pretty sure that the high nibble is important. For another example, look at packet type 0E versus 9E.) FM 23:24, 6 May 2007 (CEST)

Since you never answered, I fixed the headings. I don't have much time to work on this at the moment, so that's basically all I did. Also I'm not sure what you're talking about w.r.t packet 06 (/16?) so I left that alone. If you disagree with my analysis, that's fine, and please say so. FM 13:32, 17 May 2007 (CEST)

I also think the high nibble of the packet type is important. 0E seems to be an answer to the 00 packet when a satellite is locked, whereas 9E is one of the possible answers (9D,9E,9F) to the 05 packet. I have started to analyze the cold start trace at [1], there are some statistics about the 00 packet at [2] and a closer look on the sequence of events on a single channel (ch 1) [3]. I used a shell script [4] to filter the serial communication out of the strace output. PH5 16:19, 17 May 2007 (CEST)

Except for the flags, the packet format is now mostly understood [5]. The 'packet type' is really an initial register offset, every channel has its own set of registers, and the 0x2 flag seems to switch to a completely different register bank. PH5 20:58, 22 June 2007 (CEST)

Is there need for a SUPL-Client

Hi folks, i have started a project at jsupl.sourceforge.net to make a free implementation of the SUPL protocol. Is there any need at the openmoko project for this? Regards, Lukas