Con Kolivas
|
91edf84c60
|
Give a warning but don't reset if the avalon buffer is full early.
|
12 years ago |
Con Kolivas
|
abfaf36112
|
Discard any reads obtained from the avalon get results thread during a reset.
|
12 years ago |
Con Kolivas
|
c1dd052596
|
Differentiate initial reset in avalon from subsequent ones.
|
12 years ago |
Con Kolivas
|
589e08772e
|
Perform a mandatory reset if the avalon buffer signals it's full before it has queued its normal quota of work.
|
12 years ago |
Con Kolivas
|
b790153990
|
Wait till buffer is cleared after sending idle tasks to avalon before returning from avalon_idle.
|
12 years ago |
Con Kolivas
|
f164446282
|
Lock qlock mutex during reset from read thread in avalon to prevent more work being sent till the reset is over.
|
12 years ago |
Con Kolivas
|
499b25068c
|
Reset avalon if we continue to be unable to send all the work items.
|
12 years ago |
Con Kolivas
|
b2b594b80e
|
Revert "Don't reset result counter to zero if it's negative in avalon on just one successful share."
This reverts commit ce0a9cd3a4 .
This wasn't the problem.
|
12 years ago |
Con Kolivas
|
ce0a9cd3a4
|
Don't reset result counter to zero if it's negative in avalon on just one successful share.
|
12 years ago |
Con Kolivas
|
4c3fe39e94
|
Add avalon reset response to debugging output.
|
12 years ago |
Con Kolivas
|
d0132a7702
|
Do a wait_avalon_ready before sending a reset code.
|
12 years ago |
Con Kolivas
|
1f493d6462
|
Iterate over spare bytes in the avalon result returned from a reset request trying to find the beginning of the reset.
|
12 years ago |
Con Kolivas
|
467d149785
|
Idle avalon after reset.
|
12 years ago |
Con Kolivas
|
1ef027c444
|
Check for nothing but consecutive bad results on avalon and reset the FPGA if it happens.
|
12 years ago |
Con Kolivas
|
44736b3012
|
Make submit_nonce return a bool for whether it's a valid share or not.
|
12 years ago |
Con Kolivas
|
e90cf62af1
|
Unset the work restart flag sooner in avalon_flush_work to avoid re-entering the flush work function and just reset the queued counter instead of rotating the array to avoid runs of no valid work.
|
12 years ago |
Con Kolivas
|
0a9d7ca6b3
|
Implement an avalon_flush_work function for work restarts.
|
12 years ago |
Con Kolivas
|
3c39d75fc9
|
Shut down avalon read and write threads and idle the miners on closing it.
|
12 years ago |
Con Kolivas
|
c9ca6ca7e3
|
Tighter control over work submissions in avalon allows us to use a smaller array.
|
12 years ago |
Con Kolivas
|
db7c72b892
|
Rotate avalon array to reset the queued count before releasing the lock so work will always be available on next pass.
|
12 years ago |
Con Kolivas
|
66e27a10c5
|
Move avalon read thread start till after conditional wait, store idle status in avalon_info and use it to determine whether an error is appropriate or not.
|
12 years ago |
Con Kolivas
|
0763070cb0
|
Wait till the avalon_send_tasks thread has filled the avalon with idle work before starting the avalon_get_results thread.
|
12 years ago |
Con Kolivas
|
7cf154bfa5
|
Use AVA_GETS_OK macro in avalon_read.
|
12 years ago |
Con Kolivas
|
ebf885520d
|
Do all writes on avalon with a select() timeout to prevent indefinite blocking and loop if less than desired is written.
|
12 years ago |
Con Kolivas
|
596784aaf3
|
Check explicitly that ava_buffer_full equals the macro.
|
12 years ago |
Con Kolivas
|
e9e6ad821d
|
Send initial reset as an avalon task to remove avalon_write function.
|
12 years ago |
Con Kolivas
|
e27ca54408
|
avalon_clear_readbuf is no longer required.
|
12 years ago |
Con Kolivas
|
7dc3485fc2
|
Check for 2 stray bytes on avalon reset.
|
12 years ago |
Con Kolivas
|
629feb5f3f
|
Create a separate thread for handling all work and idle submission to the avalon which messages the scanhash function it has completed to update statistics.
|
12 years ago |
Con Kolivas
|
c87f4c303f
|
Give the avalon get results thread name the device number as well.
|
12 years ago |
Con Kolivas
|
4076e6b36e
|
Make sure we're not adjusting temps on every successful work retrieval on avalon.
|
12 years ago |
Con Kolivas
|
4784ecce75
|
Count missing work items from behind a successful work read in avalon as well.
|
12 years ago |
Con Kolivas
|
70a3e885e0
|
Change message for work not found in avalon parser.
|
12 years ago |
Con Kolivas
|
fba70ff593
|
Simplify debugging and only discard from avalon read buffer if at least one full result has been discarded.
|
12 years ago |
Con Kolivas
|
7324a33b55
|
Only display discarded bytes in avalon if they're not used as nonces.
|
12 years ago |
Con Kolivas
|
be3224f933
|
Only loop once through avalon_parse_results, but do so after timeouts as well.
|
12 years ago |
Con Kolivas
|
1f08809a7f
|
Only debug and move ram if spare bytes exist in avalon buffer.
|
12 years ago |
Con Kolivas
|
9f59e9a6d6
|
Remove off by one error.
|
12 years ago |
Con Kolivas
|
0a82b255eb
|
Inverted logic.
|
12 years ago |
Con Kolivas
|
b6b1abb5d9
|
Add more debugging to avalon reads.
|
12 years ago |
Con Kolivas
|
ab73935875
|
Convert unsigned size_ts to ints for parsing avalon messages.
|
12 years ago |
Con Kolivas
|
bb06c2f01b
|
Cope with not finding nonces in avalon parsing gracefully by not overflowing buffers.
|
12 years ago |
Con Kolivas
|
e9429f3886
|
Adjust avalon temp values on one lot of valid nonces from the parser.
|
12 years ago |
Con Kolivas
|
7404446781
|
Created a threaded message parser for avalon reads.
|
12 years ago |
Con Kolivas
|
ff44696a20
|
Avalon_wait_write is not effective during resets so do it after going idle.
|
12 years ago |
Con Kolivas
|
4794ce702e
|
Send only a single byte reset.
|
12 years ago |
Con Kolivas
|
e2c4379290
|
Repeat going idle after avalon reset, and wait for write ready before sending each reset request instead of some arbitrary sleep time.
|
12 years ago |
Con Kolivas
|
a1cec2cdaf
|
Revert "Send full avalon init task for reset."
This reverts commit fa1bca3118 .
No difference.
|
12 years ago |
Con Kolivas
|
fa1bca3118
|
Send full avalon init task for reset.
|
12 years ago |
Con Kolivas
|
c7b1b03085
|
Timeouts on avalon_read and avalon_write should be 100ms.
|
12 years ago |