1245 Commits (5bd1b560acfaced29d96a985440b4b8658d88760)

Author SHA1 Message Date
Con Kolivas 36c6da8ad4 Reinstate block solve testing. 11 years ago
Con Kolivas 3f6b9d6796 Dramatically simplify the calculation of blockdiff. 11 years ago
Con Kolivas ca91994709 Simplify the set_target function, allowing it to work properly for fractional diffs. 11 years ago
Con Kolivas 3ffc3f1d78 Adjust the binary ntime data in submit_noffset_nonce even when there is no hex ntime string for eg. gbt. 11 years ago
Con Kolivas 316e6d61c9 Put an entry into the work struct telling drivers how much they can roll the ntime themselves. 11 years ago
Con Kolivas 4728dd82d1 Remove the applog on miner threads dying to prevent deadlocks on exit. 11 years ago
Con Kolivas 4c979d756f Do one extra guaranteed libusb event handling before testing if there are any pending async usb transfers. 11 years ago
Con Kolivas 0da551fac1 Use a linked list for all usb transfers instead of just cancellable ones. 11 years ago
Con Kolivas d342bcbe81 Provide a mechanism for informing drivers of updated work templates for stratum and gbt mining. 11 years ago
Con Kolivas 9d8e3df0a8 Add usb transfer cancellation on shutdown and documentation regarding where cancellable transfers are suitable. 11 years ago
Con Kolivas d2feacdfb2 Cancel cancellable usb transfers on work restart messages. 11 years ago
Con Kolivas 448f8dde80 Don't attempt to disable curses or print a summary during an app restart to prevent deadlocks. 11 years ago
Con Kolivas 81dfe60d36 Keep the libusb event handle polling thread active until there are no async usb transfers in progress. 11 years ago
Con Kolivas e76421320e Minor opencl build corrections. 11 years ago
Con Kolivas 273b04be62 Do the ntime binary modification to the work struct when submitting an ntime offset nonce within submit_noffset_nonce 11 years ago
Roy Badami c69a130dc1 Improvements to support for BitBurner boards 11 years ago
Con Kolivas 06776af000 Use the forcelog function on shutdown to cope with indeterminate console lock states due to killing of threads. 11 years ago
Con Kolivas 8e9f32a81b Add a forcelog variant of applog which invalidates any console lock to force output. 11 years ago
ckolivas 680f014c85 Simplify queued hashtable by storing unqueued work separately in a single pointer. 11 years ago
ckolivas 40821d1f11 Use cgcompletion timeouts for the unreliable shutdown functions on kill_work. 11 years ago
Kano c9adb1929c allow __work_complete() access 11 years ago
Kano 514b3c424f API add 'MHS %ds' to 'summary' 11 years ago
Kano d3d3fc1316 Optional lock tracking and stats via the API 11 years ago
Con Kolivas 685c2acecd Speed up polling repeat again in usb poll thread and handle async after the message to disable polling is complete. 11 years ago
Kano 55e233d52c API add 'MHS %ds' to 'summary' 11 years ago
Kano c2073f0944 Optional lock tracking and stats via the API 11 years ago
Con Kolivas bec75b61a6 Make the USB polling thread poll every second to potentially aid longer timeout transfers. 11 years ago
ckolivas f821d0fbf3 Set device_diff on work in get_work to not be missed with drivers that use get_work directly. 11 years ago
Con Kolivas b6831453b4 Make the USB polling thread poll every second to potentially aid longer timeout transfers. 11 years ago
ckolivas fe5da96727 Set device_diff on work in get_work to not be missed with drivers that use get_work directly. 11 years ago
Con Kolivas 07837072f1 hexstr is too small in test_work_current 11 years ago
Con Kolivas 56edabc64d Use a timeout with usb handle events set to a nominal 200ms and wait for the polling thread to shut down before deinitialising libusb. 11 years ago
Con Kolivas 9f8023a959 Use stack memory for hex used in stratum share submissions. 11 years ago
Con Kolivas e0c90359fd Use stack memory in test_work_current, avoiding a malloc/free cycle each time. 11 years ago
Con Kolivas 835ad82441 Convert the bitfury driver to use the hash_driver_work version of hash_work. 11 years ago
Con Kolivas 58008a3742 Add a hash_driver_work function to allow for drivers that wish to do their own work queueing and management. 11 years ago
Chris Chua b509e84376 using separate identifier for bitburner fury boards 11 years ago
Con Kolivas cda797d1be Avoid calling get_statline_before on exit to avoid trying to use it on drivers in an indeterminate state. 11 years ago
Con Kolivas d437de8428 Avoid calling get_statline on exit. 11 years ago
Con Kolivas b5dfacef24 We are always dependent on libusb handling events so use the blocking libusb_handle_events in the polling thread and use a bool to know if we should continue polling. 11 years ago
Con Kolivas 2834eb8d6d Kill off threads that have failed using hash_sole_work instead of just disabling them. 11 years ago
Con Kolivas f8e28f251d Try switching pools if for some reason we end up with only idle pools and have ended up current_pool set to an idle one. 11 years ago
Con Kolivas 3b527f14d0 Check a pool is stable for >5 mins before switching back to it. 11 years ago
Con Kolivas 7318eceef9 Make the devlock a cglock in usbutils and only grab the write lock for fundamental changes allowing us to send and receive transfers concurrently without lock contention. 11 years ago
Con Kolivas 8c0ea19e13 Submitting an ntime offset nonce needs to be done on a copy of the work instead of the original so abstract out shared components as much as possible, minimising strdups in copy_work and make submit_work_async work take copied work, cleaning up code in the process. 11 years ago
Con Kolivas b6a97de687 Provide a way for drivers to submit work that it has internally rolled the ntime value by returning the amount it has ntime rolled to be added. 11 years ago
ckolivas 6cd7f0f1be Remove unmaintained broken ztex driver. 11 years ago
Kano 51418f79e5 Klondike update code to current git 11 years ago
Con Kolivas a6a9407067 Use asynchronous transfers for all bulk transfers, allowing us to use our own timers and cancelling transfers that take too long. 11 years ago
Con Kolivas 59ff31d821 Make the devlock a cglock in usbutils and only grab the write lock for fundamental changes allowing us to send and receive transfers concurrently without lock contention. 11 years ago