lupin012

Executing the `debug_traceTransaction` call using the following input: ``` { "jsonrpc":"2.0", "method":"debug_traceTransaction", "params":[ "0xeb3803f052e817e41e381b5ac150d0967e37b2e298d9a1dc079750abeeafbead", { "disableStorage": false, "disableMemory": false, "disableStack": false } ], "id":1 } ``` we see a mismatch...

For performance reason it could be better move the allocation of the EVMExecutor [outside the loop](https://github.com/torquem-ch/silkrpc/blob/master/silkrpc/commands/eth_api.cpp#L1105). Bug: In this case the tracer notify different address in the execution_state.msg->recipient for each...

Should be evaluated to reuse silkworm::ExecutionProcessor() instead of re-implement the "executor" in silkrpc in: - EvmExecutor - DebugExecutor - TraceCallExecutor

Looking at these [performance measurements](https://github.com/erigontech/silkworm/actions/runs/8740391766/job/23983963775), it is possible to spot some degradation when increasing the request rate (i.e. query-per-second/duration, QPS/T): - at rate 10000:20 we have max latency about 15...