 Luke Dashjr
|
de6e604d7c
titan: Remove unused variables
|
11 years ago |
 Pelle Nilsson
|
142bdb6705
Use new get_work_status prototype and print FPGA CRC error counters to log
|
11 years ago |
|
Pelle Nilsson
|
8e9c72dd0e
Titan flush optimization fix: Don't send get_info commands when FPGA is speaking to ASIC
|
11 years ago |
|
Pelle Nilsson
|
2582521baf
All dies use same nonce working range
|
11 years ago |
|
Pelle Nilsson
|
772a6c1197
Bugfix: First core in each die got too big nonce area to work on
|
11 years ago |
|
Pelle Nilsson
|
bfd5d25ebb
Titan flush optimization bugfix: FPGA status was checked once per poll instead of once per ASIC
|
11 years ago |
|
Pelle Nilsson
|
eb588b5007
Titan flush optimization: Bugfix: Flush time measurement only worked with one ASIC
|
11 years ago |
|
Pelle Nilsson
|
16d52ba670
Titan flush optimization: Remove temporary debug messages
|
11 years ago |
|
Pelle Nilsson
|
f567106334
Titan flush optimization: Keep works in local queue until slot number is reused
|
11 years ago |
|
Pelle Nilsson
|
4fa8c99b24
Titan flush optimization: Measure new flush time
|
11 years ago |
|
Pelle Nilsson
|
1cb4417901
Titan flush optimization: Send works to all dies after flush
|
11 years ago |
|
Pelle Nilsson
|
92e4343384
Titan flush optimization: First rough implementation
|
11 years ago |
 Vitalii Demianets
|
5db1bda83f
Titan: Less clobber on the screen: group some "failure" messages
|
11 years ago |
|
Vitalii Demianets
|
162207d2e6
Titan: Submit stale shares while doing flushes
|
11 years ago |
|
Vitalii Demianets
|
9957145695
Typo in comment
|
11 years ago |
|
Vitalii Demianets
|
0770a46363
Increase die inactivity timeout to 20 secs
|
11 years ago |
|
Vitalii Demianets
|
a2fa9d4e1e
Slightly improve some debug messages
|
11 years ago |
|
Vitalii Demianets
|
318850075e
Use multi-part batched SPI transfers for flushes (saves 2 secs on each flush)
|
11 years ago |
|
Vitalii Demianets
|
f6c6f331f7
titan: Increase queue prefill value up to 20
|
11 years ago |
|
Vitalii Demianets
|
46fc0033c8
Whitespaces
|
11 years ago |
|
Vitalii Demianets
|
d90a9b5cfb
Do not return error on RECONFIGURE command
|
11 years ago |
|
Vitalii Demianets
|
0b8548c604
Flag for fast broadcast flushes. Not enabled: DC/DCs trip off easily!
|
11 years ago |
|
Vitalii Demianets
|
59e91fa7ac
Monitor die health, reconfigure it if no shares in 10 seconds
|
11 years ago |
|
Vitalii Demianets
|
7d42273f77
Work assignment and flushing is per-die, not per-ASIC
|
11 years ago |
|
Vitalii Demianets
|
014dcae797
Set flush flag after re-configuring the die.
|
11 years ago |
|
Vitalii Demianets
|
0f30d086bc
Refactoring: intermediate variables for first_proc and repr
|
11 years ago |
|
Vitalii Demianets
|
f1ac6e85ac
Flush cores one-by-one right before reconfiguring them
|
11 years ago |
|
Vitalii Demianets
|
a6c6b24e45
Define for the broadcast core address
|
11 years ago |
|
Vitalii Demianets
|
31264eb7c2
Don't forget to setup minerloop locking ("control_request") mechanism
|
11 years ago |
|
Vitalii Demianets
|
2b79bfe93e
First clumsy attempt to reconfigure die
|
11 years ago |
