 Con Kolivas
|
2dd7d2f676
Use usbutils' own ftdi parser for avalon and the ftdir's own latency for managing timeouts since we can wait on reads with completely asynchronous reads+writes.
|
12 years ago |
|
Con Kolivas
|
8cade64912
Give avalon a rest between submissions.
|
12 years ago |
|
Con Kolivas
|
093a6302dd
Decrease the avalon read timeout to less than the time it takes to fill the tiny 64 byte buffer.
|
12 years ago |
|
Con Kolivas
|
f3d406ec72
Use usbutils' own cps function for slowing rate of usb writes on avalon.
|
12 years ago |
|
ckolivas
|
8416a4dbf1
Fix build for no libcurl
|
12 years ago |
|
Con Kolivas
|
652044d922
Check length before submitting sync transfers
|
12 years ago |
|
Con Kolivas
|
829f0687bf
Bump version to 3.7.2
|
12 years ago |
|
Con Kolivas
|
5d7c99f389
Revert "Return ETIMEDOUT regardless if we fail in cgsem_mswait since we may be waiting on it on shutdown and the return response is harmless."
|
12 years ago |
|
Con Kolivas
|
0485208342
Bump version to 3.7.1
|
12 years ago |
|
Con Kolivas
|
528eadb7fa
Update NEWS
|
12 years ago |
|
Con Kolivas
|
3dcae53e38
Return ETIMEDOUT regardless if we fail in cgsem_mswait since we may be waiting on it on shutdown and the return response is harmless.
|
12 years ago |
|
Con Kolivas
|
82a97a04f5
Avalon doesn't like losing its device data even after it's been unplugged.
|
12 years ago |
|
Con Kolivas
|
31bbaa1491
Don't free the cgpu data on shutting down avalon driver.
|
12 years ago |
|
Con Kolivas
|
ce196049fd
Clean up completely on avalon shutdown.
|
12 years ago |
|
Con Kolivas
|
a4c96d759a
Use cgsem timed waits in avalon driver to not miss any queued wake ups to account for async messages coming during a flush work.
|
12 years ago |
|
Con Kolivas
|
7011a7a008
Statline before is too long on icarus that doesn't have monitoring.
|
12 years ago |
|
Con Kolivas
|
b1be5c6d12
Different windows+usb combinations respond with varying levels of reliability wrt timeouts so use a nominal extra 40ms before cancelling transfers that fail to time out on their own.
|
12 years ago |
|
Con Kolivas
|
21f18956a8
Revert "Use no cancel transfer functions on windows in callback_wait"
|
12 years ago |
|
Con Kolivas
|
757fcb3df6
Do all hotplug_process under the write mining_thr_lock
|
12 years ago |
|
Con Kolivas
|
2489b2aeba
Fix for opt_worktime on big endian machines.
|
12 years ago |
|
Con Kolivas
|
463e366c8d
Correct set_blockdiff for big endian machines.
|
12 years ago |
|
Con Kolivas
|
043e9d9bab
Make sure cgpu exists in the restart threads loop in cases of hotplug etc.
|
12 years ago |
|
Con Kolivas
|
6bc691adb2
Treat usb write timeout errors as unrecoverable.
|
12 years ago |
|
Con Kolivas
|
42b3cf19c4
Transfer errors are filtered out in usbutils now so no need to look for them in NODEV checks.
|
12 years ago |
|
Con Kolivas
|
2e11a50484
Remove now unused entries from struct cg_usb_device
|
12 years ago |
|
Con Kolivas
|
e692cd087f
Do not double up with checking for end of timeout measurements in usb read/write.
|
12 years ago |
|
Con Kolivas
|
8f6acad8ec
Use no cancel transfer functions on windows in callback_wait
|
12 years ago |
|
Con Kolivas
|
c030439d0b
Do get_work in fill_queue without holding other locks.
|
12 years ago |
|
Con Kolivas
|
4a620cce44
Initialise usb after all the locks and conditionals are initialised.
|
12 years ago |
|
Con Kolivas
|
aa9938973e
Use only a trylock in flush queue to prevent deadlocks.
|
12 years ago |
