)]}'
{"/PATCHSET_LEVEL":[{"author":{"_account_id":597,"name":"Pete Zaitcev","email":"zaitcev@kotori.zaitcev.us","username":"zaitcev"},"change_message_id":"9e348814f1c3eb776c0ba9cea5471481a3f4062c","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"c19313c1_ecb49bc5","updated":"2025-06-11 04:08:55.000000000","message":"I\u0027m not landing this immediately because of the TODO marker. Maybe someone wants to look deeper?","commit_id":"4044849b1e519f79cfbfb7fa282549b32a224132"},{"author":{"_account_id":7233,"name":"Matthew Oliver","email":"matt@oliver.net.au","username":"mattoliverau"},"change_message_id":"7c03b69b071dd37d91163f69cc2db29e918d7a82","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":1,"id":"3e11bd75_8b9897b3","updated":"2025-06-13 00:57:00.000000000","message":"recheck","commit_id":"4044849b1e519f79cfbfb7fa282549b32a224132"},{"author":{"_account_id":34930,"name":"Jianjian Huo","email":"jhuo@nvidia.com","username":"jhuo"},"change_message_id":"a7ef0da2730fbfff20871cd8ce4da1a2e5f77208","unresolved":false,"context_lines":[],"source_content_type":"","patch_set":2,"id":"43d035b4_1cabf091","updated":"2026-01-20 20:29:24.000000000","message":"linux kernel’s default socket.SO_RCVBUF is probably larger than our previous default 64KB client_chunk_size typically on a lot of platforms, so essentially we are using kernel receive buffer with py3.\n\nI feel the current situation is fine, ideally operator should align client_chunk_size with socket.SO_RCVBUF default size though. so that\u0027ll be great for this patch to add a comment of this at here: https://github.com/NVIDIA/swift/blob/master/swift/proxy/controllers/obj.py#L1013-L1019","commit_id":"8ee74fcc1d4f8f55fd95fb09f1f49dba299f9160"}]}