01:27 <DocScrutinizer05> apelete: dunno if that's interesting for you: http://processors.wiki.ti.com/index.php/OMAP35x_To_AM37x_Software_Migration_Guide#USB_Drivers

01:28 <DocScrutinizer05> Mr Balbi might have missed on a few details there

09:07 <apelete> good morning

09:08 <apelete> DocScrutinizer05: interesting indeed, I guess I'll have to be aware of changes in the musb otg driver from now on

09:08 <apelete> thanks for the link

09:09 <apelete> larsc: noticed something very strange yesterday in glue layer code

09:10 <DocScrutinizer05> yw

09:11 <apelete> larsc: if I enable debug messages with #define DEBUG, hnp bug still occurs -> http://paste.debian.net/66919/

09:11 <apelete> larsc: but without #define DEBUG, everything seem fine -> http://paste.debian.net/66917/

09:12 <apelete> any idea to explain this ?

09:14 <apelete> looked into it until very late last night, don't understand what's going on

09:18 <apelete> maybe I should try with kernel 3.12 instead of 3.11

09:19 <larsc> most likely one of the debug printks does a NULL pointer dereference

09:21 <apelete> I really hope you're right, truth is I couldn't think straight past 2AM last night

09:22 <apelete> anyway, will try to look into it tonight

09:24 <apelete> larsc: on a brighter side, let's say it's a good reason for me to push jz-3.12 as soon as possible and continue hacking from there ;-)

11:55 <wpwrak> larsc: do you know a bit about the ADF7242, a RF+MCU chip, apparently suitable for 802.15.4 and BT ?

11:57 <larsc> my boss wrote the linux driver for it

11:58 <larsc> I think

11:58 <larsc> but that's all I know

11:58 <wpwrak> so it's one of those chips. very good :)

11:58 <larsc> it's the only one ;)

11:59 <larsc> https://wiki.analog.com/resources/tools-software/linux-drivers/networking-mac802154/adf7242

12:00 <wpwrak> it came up as a potential BT solution for anelok: all the things that say "BT" tend to have their share of openness issues

12:00 <wpwrak> but there's a small number of RF chips that seem to be able to do BT-style modulation

12:00 <wpwrak> and there's a project that's slowly working towards having an all-open BT stack: ubertooth

12:01 <wpwrak> they're currently using the TI CC2400. that one's about to EOL (not recommended for new designs)

12:01 <wpwrak> and its successor(s) don't have the same capabilities

12:02 <wpwrak> but they do think the ADF7242 may be a potential future replacement. not sure how far they went with their evaluation, though.

12:04 <whitequark> wpwrak: have you considered a BLE-only device?

12:05 <whitequark> BLE is way simpler and there seem to be more devices supporting it, e.g. nRF24L01+ almost cuts it

12:05 <wpwrak> i'm not sure how the two are really related when it comes to RF

12:05 <whitequark> well

12:05 <wpwrak> do you know the modulation characteristics ?

12:05 <whitequark> BLE and BT share the name. that's about the only thing they share.

12:05 <whitequark> yeah

12:05 <wpwrak> yeah :)

12:05 <whitequark> let me find an article

12:05 <whitequark> . The frequency is 2.4GHz, chanels are 1MHz apart, modulation is GFSK, datarate is 1MBps, preamble is 10101010 or 01010101 based on first data byte, and addressing is done using a 32-bit address.

12:07 <wpwrak> BTLE seems to channel a lot of 802.15.4. they even happily went ahead and duplicated a good part of the linux kernel 802.15.4 stack, not realizing that it what they copied was LOADED with bugs (but they took enough care to edit the headers to change the authorship ;-)

12:07 <wpwrak> well, all of this has been corrected now ...

12:07 <whitequark> http://dmitry.gr/index.php?r=05.Projects&proj=15&proj=11.%20Bluetooth%20LE%20fakery

12:08 <wpwrak> 2.4 GHz band, 1 MHz spacing, GFSK at 1 Mbps sounds like regular BT 1.2

12:10 <whitequark> dmitry says this might work better: http://www.elecfreaks.com/wiki/index.php?title=2.4G_Wireless_easy_radio_A7125

12:18 <wpwrak> that A7125 looks like a darling to source. not even octopart has heard of it.

12:20 <wpwrak> and it's FSK, not GFSK

12:21 <wpwrak> of course, if elecfreaks sell a whole module for USD 2.95, the chip can't be too expensive. i like that :)

12:24 <wpwrak> A7105 looks better

12:27 <wpwrak> sourcing would be aliexpress. i shuold learn not to be surprised :)