tevador

Yes, it would be great if you could do the changes in the ARM64 JIT. But please wait, I realized the JitCompiler interface needs to be changed because the class...

This is the offending input: `0707c7b1f0d60567d143dd0d0b600a54ed14169486b016e77266cc63bf9a79b96f9bcadd4de2c37201000052a04939ea39bfd3c8595707bd8fe4b1e7d8046ca88bfee4dc5483ad37e5758103` (nonce = 370) It seems the nested for loops make it run for hours. The generated program can be found here: https://pastebin.com/9YimzA3N

@livinginformation Sounds similar to my concept [CNVM](https://github.com/tevador/cnvm). I think the only way to make the runtime approximately equal between different programs is to execute a fixed number of instructions/statements.

I think the simplest solution is to just throw away the nonce (kill the execution) if the program runs for more than (mean + x * sigma). It's probably acceptable...

Full nodes must download the whole blockchain in order to verify it. There is currently no way around that (perhaps in the future there can be a more efficient way...

If verification is skipped for checkpointed blocks, that is unfortunate and should be fixed ASAP. The default should be full verification of the blockchain. There can be an option to...

> when both fast sync and pruning are enabled simultaneously, --sync-pruned-blocks should also be enabled by default Yes, I agree. Downloading data to be discarded without being verified is just...

`--fast-sync` is a hypothetical command which does not exist. Fast sync is on by default but can be disabled with `--fast-block-sync 0`. `--sync-pruned-blocks` doesn't work if fast sync is disabled.

The biggest issue with moving to another curve would be the required migration of amount commitments (in order to prove that transactions balance out). Any transaction spending RingCT e-notes (and...

Intel Xeon E3-1245 @ 3.7 GHz (virtual machine): `1000 nonces took 7.860379 seconds` Cryptonight hashrate = 71 H/s per thread Intel Core i3-3220 @ 3.3 GHz: `1000 nonces took 9.508519...