Gregory Nutt
|
c9661ad5a7
|
Change all time conversions. Yech. New timer units in microseconds breaks all existing logic that used milliseconds in the conversions. Something likely got broken doing this, probably because I confused a MSEC2TICK conversion with a TICK2MSEC conversion. Also, the tickless OS no appears fully functional and passes the OS test on the simulator with no errors
|
2014-08-07 18:00:38 -06:00 |
|
Gregory Nutt
|
17ea50b45e
|
Fix compler error in tickless code when round robin scheduling is enabled. Other cosmetic fixes to comments
|
2014-08-07 14:14:09 -06:00 |
|
Gregory Nutt
|
822a4fafeb
|
Update ChangeLog
|
2014-08-07 13:43:56 -06:00 |
|
Gregory Nutt
|
0aa7209765
|
Change CONFIG_MSEC_PER_TICK to CONFIG_USEC_PER_TICK. This gives more options for system timers in general, but more importantly, let's us realize higher resolution for the case of CONFIG_SCHED_TICKLESS=y -- of course, at the risk of some new interger overvflow problems
|
2014-08-07 13:42:47 -06:00 |
|
Gregory Nutt
|
6bdc08226f
|
Update ChangeLog
|
2014-08-07 12:40:16 -06:00 |
|
Gregory Nutt
|
caba61999a
|
Remove CONFIG_DISABLE_CLOCK
|
2014-08-07 12:35:24 -06:00 |
|
Gregory Nutt
|
a2de3c92d8
|
Update ChangeLog
|
2014-08-07 11:40:54 -06:00 |
|
Gregory Nutt
|
520a51a3e1
|
Implements the tickless OS
|
2014-08-07 11:39:16 -06:00 |
|
Gregory Nutt
|
317b66ca37
|
Update ChangeLog
|
2014-08-06 18:33:44 -06:00 |
|
Gregory Nutt
|
dc48bec019
|
Add support for a simulated interval timer support verification of the tickless OS.
|
2014-08-06 18:29:29 -06:00 |
|
Gregory Nutt
|
aaafcfd688
|
if CONFIG_SCHED_TICKLESS is defined, then the global variable g_system_timer does not exist.
|
2014-08-06 18:29:05 -06:00 |
|
Gregory Nutt
|
a1ae9d36df
|
Don't build in sched_processtimer.c if CONFIG_SCHED_TICKLESS is selected.
|
2014-08-06 18:28:51 -06:00 |
|
Gregory Nutt
|
c8a60b2916
|
Added another tickless OS interface to get the timer remaining on the interval timer.
|
2014-08-06 18:28:33 -06:00 |
|
Gregory Nutt
|
b0324c00ee
|
Uses interval timer interfaces to get the time if CONFIG_SCHED_TICKLESS is selected.
|
2014-08-06 18:28:08 -06:00 |
|
Gregory Nutt
|
a93fdc165d
|
Use the clock_systimer() macro, do access the g_system_timer() global directly.
|
2014-08-06 18:27:35 -06:00 |
|
Gregory Nutt
|
cc7cb0f031
|
Don't build in sched_processtimer.c if CONFIG_SCHED_TICKLESS is selected.
|
2014-08-06 18:27:10 -06:00 |
|
Gregory Nutt
|
0452b1555b
|
if CONFIG_SCHED_TICKLESS is defined, then the global variable g_system_timer does not exist
|
2014-08-06 18:26:16 -06:00 |
|
Gregory Nutt
|
f0eb4235d1
|
Add a new option CONFIG_SCHED_TICKLESS to enable the tickless OS. This is a work in progress but should be relatively safe since the new features are only enabled by this option (Safer but not 100% safe since it does not protect from typos and other dumb errors).
|
2014-08-06 18:25:42 -06:00 |
|
Gregory Nutt
|
e1f9786db8
|
Update ChangeLog
|
2014-08-06 16:29:46 -06:00 |
|
Gregory Nutt
|
736d3c169a
|
Rename up_timerinit() to up_timer_initailize(); Add prototypes for candidate interfaces for the tickless OS; Don't build existing timer initialization logic if CONFIG_SCHED_TICKLESS is defined.
|
2014-08-06 16:26:01 -06:00 |
|
Gregory Nutt
|
5e045c26c2
|
Updated README
|
2014-08-06 12:30:35 -06:00 |
|
Gregory Nutt
|
0e134b5233
|
Update some defconfig files; enable network initialization thread
|
2014-08-06 12:22:51 -06:00 |
|
Gregory Nutt
|
f75b9c8e99
|
Update TODO and ChangeLog
|
2014-08-06 12:20:38 -06:00 |
|
Gregory Nutt
|
c078814d98
|
NSH networking: There is now a configuration option that will bring up the network on an separate thread. Since the network bring-up is asynchronous, there are not serial console start-up delays due to the network negotiation time.
|
2014-08-06 11:59:41 -06:00 |
|
Gregory Nutt
|
b33beb3e20
|
NSH Networking: Also add an option to let platform-specific logic select the MAC address.
|
2014-08-06 10:12:10 -06:00 |
|
Gregory Nutt
|
0ae6c68a01
|
Update ChangeLogs
|
2014-08-05 10:09:43 -06:00 |
|
Gregory Nutt
|
78ccbdeb6e
|
PCM decoder: Handle errors in the PCM WAV file in a way that the Nxplayer can recover gracefully: Send and Audio Complete message.
|
2014-08-05 10:06:05 -06:00 |
|
Gregory Nutt
|
bb29ba0a00
|
Costmetic changes
|
2014-08-05 10:03:29 -06:00 |
|
Gregory Nutt
|
6e71fa30f1
|
Update ChangeLog
|
2014-08-05 07:07:52 -06:00 |
|
Gregory Nutt
|
7b9c44101d
|
SAMA5D3 HSMCI: TX DMA is again disabled
|
2014-08-05 07:07:39 -06:00 |
|
Gregory Nutt
|
73db3d052f
|
Update ChangeLog
|
2014-08-04 19:29:02 -06:00 |
|
Gregory Nutt
|
8065ab73a4
|
WM8904: Add reset logic to put the part back in its initial state after playing each WAV file. Base samles per second on frame length, not bits-per-sample. Use a different frame length for 8-bit and 16-bit data
|
2014-08-04 19:26:43 -06:00 |
|
Gregory Nutt
|
23024b2d6a
|
SAMA5: Some usage documents for media players
|
2014-08-04 19:23:20 -06:00 |
|
Gregory Nutt
|
cc5c1b6539
|
WM8904: Update READMEs and defconfigs to last tested configuration
|
2014-08-04 15:36:16 -06:00 |
|
Gregory Nutt
|
25b58a4241
|
Update ChangeLog
|
2014-08-04 15:12:34 -06:00 |
|
Gregory Nutt
|
dc8b040573
|
WM8904: Correct calculation of bitrate (I am not sure why this is correct). LR clock divisor is now a constant 32-bits per frame. Conditioned out interrupt logic; it is not being used. Also added some FLL lock interrupt/poll logic (which was subsequently disabled).
|
2014-08-04 15:08:20 -06:00 |
|
Gregory Nutt
|
e30d31367a
|
WM8904 interface enable method now returns the previous interrupt state. Correct ordering of some WM8904: Need to provide MCLK before initializing the WM8904, not after
|
2014-08-04 14:56:20 -06:00 |
|
Gregory Nutt
|
eebdae30b2
|
Correct the WM8904 interrupt configuration, active high level. Fix that and don't pull-up
|
2014-08-04 14:54:20 -06:00 |
|
Gregory Nutt
|
b305f9186a
|
WM8904 interface enable method now returns the previous interrupt state
|
2014-08-04 14:52:59 -06:00 |
|
Gregory Nutt
|
c68e42c437
|
WM8904: Restore mystery bit in setup that was accidentally tossed before
|
2014-08-03 19:29:02 -06:00 |
|
Gregory Nutt
|
678c0aac8d
|
WM8904: Tried disabling the SYSCLK while updating the FLL. Didn't help but is still probably a correct change
|
2014-08-03 18:44:51 -06:00 |
|
Gregory Nutt
|
72b739fbcb
|
WM8904: Add logic to analyze BCLK setup
|
2014-08-03 17:11:38 -06:00 |
|
Gregory Nutt
|
b9384ced60
|
WM8904: Various updates to get BCLK/LRCLK correct. Acutally worse the parent, but I think is on the right track
|
2014-08-03 13:28:20 -06:00 |
|
Gregory Nutt
|
c948521408
|
Update ChangeLog
|
2014-08-03 11:02:42 -06:00 |
|
Gregory Nutt
|
1f213064e0
|
SAMA5 WM8904: Add option to use MAINCK as WM8904 MCLK
|
2014-08-03 11:01:02 -06:00 |
|
Gregory Nutt
|
55e6fd5381
|
Update ChangeLog
|
2014-08-03 10:19:26 -06:00 |
|
Gregory Nutt
|
159bcc255d
|
SAMA5 PCK: Add Main clock as an option for the PCK clock source
|
2014-08-03 10:17:50 -06:00 |
|
Gregory Nutt
|
eafa4aeaa1
|
Update ChangeLog
|
2014-08-02 14:30:46 -06:00 |
|
Gregory Nutt
|
36f8e23c12
|
WM8904 driver: Fix wm8904_dump_registers macro: Wrong number of parameters.
|
2014-08-02 14:28:30 -06:00 |
|
Gregory Nutt
|
b0d0ac72a7
|
WM8904 driver: Disable the FFL at the end of playing.
|
2014-08-02 14:28:10 -06:00 |
|