Avi Kivity
Avi Kivity
Using submit_to() would result in a lot of delay. If we tolerate delay, we can just randomize the timer.
How about: shard 0 formats and stores in new memory. Old memory is deallocated after we're sure that other shards no longer use: ``` std::atomic _buffer; char* new_date = ...;...
@vladzcloudius ?
This can happen if the NIC uses a hash function different from what we think it is. What NIC are you using? Please test with --smp 1, just to validate.
Try changing `if (smp::count > 1)` to `if (false)` in `dpdk_device::init_port_start()`. If it works, we know it's a hash function mismatch. /cc @vladzcloudius
This looks like a ceph failure, not seastar. Of course it can be caused by a seastar bug, but it's not possible for me to diagnose ceph assertion failures. Maybe...
client connect()s have extra hashing logic: ```c++ template auto tcp::connect(socket_address sa) -> connection { uint16_t src_port; connid id; auto src_ip = _inet._inet.host_address(); auto dst_ip = ipv4_address(sa); auto dst_port = net::ntoh(sa.u.in.sin_port);...
Dependency: https://github.com/scylladb/seastar/pull/2903
Rebased past 4ce3e66f0b42bd4fc7a54db367ca79918da69385
v3: rebased past fixes (719d7880eca70d7be3e25c10f41a01656ea5278c, b1a7ed56183a6f414bd8f654394b31f89245bda3)