bunnie
08a86749fe
fix hosted mode bugs
2021-05-09 01:13:16 +08:00
bunnie
a8a35fa434
roll back heapless crate on ime-frontend
2021-05-08 23:55:08 +08:00
bunnie
a24803354c
refactor xous-names to eliminate heapless
...
see issue #30
2021-05-08 23:34:08 +08:00
bunnie
391122823d
refactor heapless version in attempt to debug issue that cropped up in xous-names
...
somehow, updating the toolchain has broken the FnvIndexMap implementation
2021-05-08 18:44:23 +08:00
bunnie
b904a114fe
this commit crashes on boot
2021-05-08 16:40:44 +08:00
bunnie
7f7bdbff88
this will play the long sample. and it works! :)
2021-05-08 03:41:04 +08:00
bunnie
15fc4a6c36
audio working, trying out longer samples
2021-05-08 03:14:14 +08:00
bunnie
fe6b417643
checkpoint
2021-05-08 01:13:30 +08:00
bunnie
ab67cf46b2
fix path to renode
2021-05-07 23:43:52 +08:00
bunnie
46fa8fa01e
extend shellchat to allow for multiple callback registrations
2021-05-07 16:22:06 +08:00
bunnie
b3a60faec2
audio checkpoint
2021-05-07 15:09:51 +08:00
bunnie
7dd66a4c18
playback is mostly working and stable
2021-05-07 05:42:39 +08:00
bunnie
adc7f8e98d
cleanup the i2c API
...
condense async/sync methods into a single one and decide which
is which based upon the presence of an Option<callback>
2021-05-06 14:03:06 +08:00
bunnie
b50aa60591
clean up some visibilities and warnings, get rid of the old, bad API
2021-05-06 05:13:23 +08:00
bunnie
bc82dbd157
transition i2c api on codec, finally found the damn race condition
2021-05-06 05:05:17 +08:00
bunnie
75bb4805b1
clean up RTC server to use the new I2C API
2021-05-06 04:16:31 +08:00
bunnie
b596c40dd6
refactor I2C to have work queue, cleaner (but untested) APIs
2021-05-06 03:40:04 +08:00
bunnie
bb5bc1d212
refactor i2c to break out the i2c machine to a separate server
...
the concurrency issues on I2C are getting hairy, and it was probably a
dubious decision from the start to merge I2C into the LLIO block; but
originally this was motivated by the idea of keeping all the
"miscellaneous slow stuff" in one process to keep the memory footprint
small and make code a bit more readable.
Keeping I2C in the same process space as the LLIO but splitting it
into its own thread and server splits the difference, by giving us
more options to deal with the concurrency, without inflating the size
of the kernel binary and allocating a lot of extra static data.
2021-05-06 00:45:14 +08:00
bunnie
c0d72c282b
increase process limit
2021-05-06 00:44:32 +08:00
bunnie
fe86207469
make docs reflect the error message actually thrown
2021-05-06 00:41:38 +08:00
bunnie
f975884f1e
increase server limit to 128
2021-05-06 00:41:27 +08:00
Sean Cross
23e759ffd8
kernel: fix multi-page lending
...
Due to several errors, multi-page lending did not work properly.
To begin with, if a page was not allocated then it would not be able to
be sent. That is, if you allocated 8192 bytes but only ever accessed the
first 4096 bytes, then an error would be encountered when lending the
unaccessed upper page.
Furthermore, due to a striding error in accessing pointers, half-pages
were treated as full pages.
Signed-off-by: Sean Cross <sean@xobs.io>
2021-05-05 10:58:12 +08:00
bunnie
ea06df6879
test case for >1 page lends not working
2021-05-05 01:37:25 +08:00
bunnie
e7e7711f2d
remove unused buffer package
2021-05-05 01:37:03 +08:00
bunnie
d17ae106f1
kernel panic on memory lend
2021-05-05 00:58:48 +08:00
bunnie
7a7e77d346
intermediate checkin, taking a break
2021-05-04 19:08:56 +08:00
bunnie
61611caddd
some general ideas to debug later
2021-05-04 02:47:03 +08:00
bunnie
f8979d6341
intermediate progress on audio. doesn't compile.
2021-05-04 02:22:35 +08:00
bunnie
a4ca725c2a
add a reset to an i2c core
2021-05-03 21:24:37 +08:00
bunnie
01e2c7edca
migrate to an interrupt-handler only I2C handler
...
the amount of time it takes to send messages to initiate each
packet is too slow, and causing timeouts. By handling the I2C
directly in an interrupt handler, we accelerate the bus, at
the expense of no visibility via log/trace messages.
2021-05-03 18:53:17 +08:00
bunnie
d65e91f792
refactor i2c to be in the implementation block
2021-05-03 18:14:58 +08:00
bunnie
8e8b515a36
add a script to provision audio artifacts
2021-05-03 18:02:20 +08:00
bunnie
50174f99e5
re-org the codec crate so that it can accommodate gnarly backends
...
This is in anticipation of:
1) supply chain issues causing CODEC chips to change out
2) host integration being a PITA
2021-05-03 00:31:05 +08:00
bunnie
f558f75116
add the skeleton for the codec server
2021-05-03 00:20:51 +08:00
bunnie
dcb9956bcc
add a "skeleton" server which is a template for other servers
2021-05-03 00:19:45 +08:00
bunnie
4854cc98ee
expand README
2021-05-02 18:15:29 +08:00
bunnie
d936b69c14
merge credits into README
2021-05-02 18:13:24 +08:00
bunnie
ba729bd172
rename as generics
2021-05-02 18:11:31 +08:00
bunnie
4fbee9cabd
add a note about the audio samples
2021-05-02 18:08:34 +08:00
bunnie
c3e02cca0e
add audio samples for testing + credits
2021-05-02 18:04:13 +08:00
bunnie
abf308e82d
try to trace down issue #61
2021-05-02 16:39:08 +08:00
bunnie
a35392f35b
Merge pull request #62 from betrusted-io/bunnie-dev
...
Bunnie dev
2021-05-02 16:14:19 +08:00
bunnie
a2280639d9
forgot to uncomment a line
2021-05-02 16:13:06 +08:00
bunnie
05df98d1c1
disable suspend/resume support in hosted mode
...
hosted mode doesn't have suspend/resume support, and it's
not worth debugging subtle issues on that side of the fence
that seem to be OS-dependent.
2021-05-02 16:10:51 +08:00
bunnie
0621fed5d6
back up debugging statements
2021-05-02 12:51:41 +08:00
bunnie
8b29fc4dd3
make the sleep note screen a little more aesthetic
2021-05-02 03:57:40 +08:00
bunnie
0ca2f58c6e
add a note to indicate if we're in cold sleep
...
it's not terribly aesthetic right now, but we can fix that later.
2021-05-02 03:36:54 +08:00
bunnie
b17753ce8b
retrofit the rest of the servers for suspend/resume
...
we can now do suspend/resume in the xous chat!
2021-05-02 01:38:44 +08:00
bunnie
85d78439d7
fcc-agent is now deprecated
2021-05-01 21:01:46 +08:00
bunnie
c7f1fe0817
cleanup after debugging ticktimer
...
the problem was that the load signal was not being respected
by the hardware, small patch to FPGA required but now all is OK.
2021-05-01 20:56:12 +08:00
