Commit graph

315 commits

Author SHA1 Message Date
Mariano Alvira e06e2143d7 handle empty COBJS 2010-03-11 16:17:18 -05:00
Mariano Alvira a61deb78f4 add a way to disable FCS checking. 2010-03-11 15:18:29 -05:00
Mariano Alvira da7898557d fix how board.a is created and linked 2010-03-11 12:22:47 -05:00
Mariano Alvira 3cd4bf5768 clear the offset when freeing a packet 2010-03-10 17:59:04 -05:00
Mariano Alvira a5f8b58e07 debug PRINTF fixes 2010-03-10 15:35:13 -05:00
Mariano Alvira ecebd094bf fix radio_off for interrupts 2010-03-10 15:35:04 -05:00
Mariano Alvira 680a028799 turn on -Werror 2010-03-10 11:21:05 -05:00
Mariano Alvira 6fa57c8e9f remove unused variable 2010-03-10 11:20:39 -05:00
Mariano Alvira 017172577c make an empty board.a
this fixes errors when trying to build board.a when COBJS is empty.
2010-03-10 11:19:43 -05:00
Mariano Alvira 1d1317a136 trim values for the production econotags 2010-03-09 18:46:37 -05:00
Mariano Alvira 67aed03b1d added interrupt based uarts. 2010-03-09 18:23:40 -05:00
Mariano Alvira db145ed8a4 removed some stale comments 2010-03-08 18:45:26 -05:00
Mariano Alvira 1ca5cf698a did something sensible to the leds
The buffered packets made the indirect indicators misleading --- now
they match.
2010-03-08 18:39:08 -05:00
Mariano Alvira f64e689d2c set the third bit to avoid acks. 2010-03-08 18:27:52 -05:00
Mariano Alvira 650e9e7f2f faster transmit delay 2010-03-08 18:13:48 -05:00
Mariano Alvira 3178906b43 surpress warning. 2010-03-08 18:13:19 -05:00
Mariano Alvira eed8544a0d rx callback 2010-03-08 18:12:59 -05:00
Mariano Alvira ea2646e5ec implement rx callback 2010-03-08 18:12:31 -05:00
Mariano Alvira 8b73c61175 change the receive timeout to be 32 large packets 2010-03-08 17:39:04 -05:00
Mariano Alvira 761d2c4cbf handle case when buffers are full and we need to transmit.
this is a problem because there won't be a buffer available for the
acknowledgment. In this case use the dummy_ack packet which is always
free for this purpose.
2010-03-08 17:37:35 -05:00
Mariano Alvira d097647edd turn on tx_on 2010-03-08 17:36:42 -05:00
Mariano Alvira 0f9f67d002 add dir set and reset 2010-03-08 17:35:34 -05:00
Mariano Alvira 6ab2c0e69a don't send ack's in rftest-tx 2010-03-08 15:22:37 -05:00
Mariano Alvira 512ddc3e87 Merge branch 'master' of git://git.devl.org/git/malvira/libmc1322x
pulled in the new maca driver.

gets 238.5 kbps sustained with 125 byte payloads (133 bytes total)

and 194.9 kbps with 16 byte payloads (24 bytes total)

The difference is speeds is due to the overhead in the handlers and
such which is at 88 us right now but can probably be improved.

Conflicts:

	TODO
2010-03-08 14:51:24 -05:00
Mariano Alvira 36f213017e use thumb again. 2010-03-08 14:49:58 -05:00
Mariano Alvira 06e3e0b207 add maca_init()
changed tests to use it.
2010-03-08 14:49:31 -05:00
Mariano Alvira 386f190720 first byte can't be anything you want it to be. 2010-03-08 12:42:37 -05:00
Mariano Alvira 362cfe43eb add starting maca_isr call.
need to put all this into maca_init()
2010-03-08 12:30:25 -05:00
Mariano Alvira 571861c6ea post_receive still needs to timeout if it can't find any free
bufferes.
2010-03-08 12:29:49 -05:00
Mariano Alvira 2a07cbe665 enabled softclk timeouts
fixed post interleaving problem

fixed interrupts in ResumeMACASync.

This seems quite solid now.
2010-03-08 12:03:20 -05:00
Mariano Alvira d3f6ef07f2 sort out the length details. 2010-03-07 18:48:47 -05:00
Mariano Alvira fb922c0ab9 transmit at max power 2010-03-07 18:48:36 -05:00
Mariano Alvira e8eaece25a fix print_packet so that it doesn't break everything 2010-03-07 17:28:40 -05:00
Mariano Alvira fece5269c7 get the newlines straight in packet_print 2010-03-07 17:14:00 -05:00
Mariano Alvira a04a69480b add the 4 byte preamble 2010-03-07 17:07:54 -05:00
Mariano Alvira 024448e65e add an offset to packet_t to compensate for the added length byte on
reception
2010-03-07 17:04:30 -05:00
Mariano Alvira 8421031d4a condense print_packet output 2010-03-07 16:50:32 -05:00
Mariano Alvira 667b238148 rftest-tx and refine post_tx for fast transmission 2010-03-07 16:49:57 -05:00
Mariano Alvira 4b17c144d0 LEDs now indicate bit position (instead of (1 << pos)) which is more
consistent with how everything else works
2010-03-07 15:39:56 -05:00
Mariano Alvira c218d1bffa update rftest-rx to use new maca. 2010-03-06 21:47:22 -05:00
Mariano Alvira 0dac24e1eb this is pretty close to working together. 2010-03-06 19:54:55 -05:00
Mariano Alvira e8550f26ba this receive seems to work well.
It has checksum errors at 123 byte payloads, but this was tested with
the old rftest-tx --- so the transmit side could be the problem.
2010-03-06 19:14:55 -05:00
Mariano Alvira f5391e88dc commented the data indication printf 2010-03-06 18:50:47 -05:00
Mariano Alvira cca5e57261 add timestamps someday 2010-03-06 18:49:08 -05:00
Mariano Alvira 2d68c30934 this receive works 2010-03-06 18:48:13 -05:00
Mariano Alvira 0f400b4341 lots of fixes.
major fix is adding a completion counter based on packet length to
restart the MACA when it freezes on transmit.
2010-03-06 17:58:49 -05:00
Mariano Alvira 5c20312744 this does 127 B packets as fast as possible (with these settings).
Transmitts are every 5.8ms and are 4.48ms long
2010-03-05 19:25:38 -05:00
Mariano Alvira 8e3c2a8622 added locking and interrupt disables.
transmit looks pretty good.
2010-03-05 17:06:08 -05:00
Mariano Alvira 0c3947ddd6 forgot to add packet.h 2010-03-05 06:59:51 -05:00
Mariano Alvira 3955bbd355 more TODO items 2010-03-04 20:50:16 -05:00