Merely compiling Bitcoin Core with thread_local seems to correlate with massively degraded performance for both the RPC and P2P interface on low-end machines. See for example:

• gitian vs self-compiled on ARM 32-bit gettxoutsetinfo #18538 (comment)
• Maybe Sync is very slow or such an error #14669
• Maybe Outgoing P2P connections are reset immediately #17247

If thread_local really is the source of the performance degradation, it should be disabled by default and only enabled by --enable-debug (for use in lock order debugging).