Mariano Sorgente 8a028c3594 Ms.mempool locking (#9050) ... * Prority locking to consensus * Remove pstats * Linting * Do some stuff outside of lock * Fix startup * Add log timings * Try some different locking * Add limit * catch excp * CLVM inside lock * Try using a semaphore instead * use events for lock queue * test * Add logging for message types * type * remove seed * check new peak waiters * correct FullNodeAPI self.full_node.new_peak._waiters typo * correct logging string typos * only warn about new_peak Waiters if there is at least 1 * remove no-longer-accepted parameter to FullNode.peak_post_processing() * only warn about respond_transaction Waiters if there is at least 1 * lint * Change some constants * Small fix and logging changes * Put message types outside * Change some log levels so we can test with info * More logging * Increase rate limits but decrease paralelism * tweaks * Log dropped tx * Fix pool rpc test * Test fixes * Mempool optimization * Remove from seen if fails * Increase queue sizes * Message types info * More test and logging * Small changes to networking just in case * Decrease logging * Decrease logging even further * Decrease logging even further even further * Decrease logging 3 * Transaction queue * Don't cancel tasks or close connection * Cancel tasks on disconnect (for shutdown purposed) * Fix typo * Catch cancelled * Do multiple at a time * More accurate farmer response time * More efficiently create tasks * Increase queue size and priority by fee * Revert priority * Don't re-request too many times for dropped TX * Handle cancelled error so we don't go into a bad state * Catch cancelled in syncing tasks * Reduce new_peak_sem to improve performance * Less bytes conversion * Missing file, and 2 workers for CLVM * Validate BLS in a new thread * tests * Change semaphore constants * correct a cancellation triggered exception and assertion * Fix send_transaction, dont use BaseException, fix tests * Fix more tests * only log transaction handler cancellation in debug * typing in log * move unfinished validation to diff proc * it is asyncio.CancelledError * Add a test for bad signature * Fix more tests, reduce logging, lint * One more lint * blockchain tests, pass bytes directly, single call * Try to fix rl_wallet failures * Fix mempool test * catch everything * Don't test RL wallet * Fix more tests and return error code * Improve error handling in multiprocess * Add pre-validation time * Add pre-validation time in logs, and revert pytest.ini changes * Add log correctly * Ms.bls cache experiment (#9115) * Logging for cache * Less logging * Return to original plan * Clean up * Remove coment * Remove log * formalize LockQueue shutdown * Comments * Fix blockchain test * Improve cache * Remove logs * Fix sign_coin_spends * Fix pool wallet Co-authored-by: Kyle Altendorf <sda@fstab.net> Co-authored-by: Yostra <straya@chia.net>		2021-11-04 09:29:05 -07:00
..
singleton.py	Pools mainnet (#7047)	2021-06-29 14:21:25 -07:00
spend_sim.py	Ms.mempool locking (#9050)	2021-11-04 09:29:05 -07:00