maflcko
maflcko
> build/out is wiped every time. Would be awesome if that was not the case, if there was, for example, an output directory per descriptor (or make the cleaning optional,...
There seems to be a scale issue with office7.obj being larger by a factor of 100 in each dimension. This will cause problems when running an algorithm over all files...
I was wondering if this library supports `-D_GLIBCXX_DEBUG_PEDANTIC` builds. Currently I am seeing a failure. Steps to reproduce: ``` $ cat 2.cpp #include #include int main() { std::string hex("31393730300b"); std::string...
Python itself is available on windows: See e.g. https://github.com/chaincodelabs/bitcoinperf/commits/1911-ciWindows However, the compilation would have to be done in `wsl`, I guess.
For some benchmarks it would be nice to measure the consumed CPU time. E.g. * ibd/reindex, which could be dominated by disk/net access. * An "idle test", where bitcoind in...
We should at least run against the version of gcc that the gitian builds are built with + the versions in LTS Ubuntu
Bitcoinperf should run a mainnet node with debug=1 (maybe exclude leveldb et al) and have the debug log visualized. Plots could show e.g. the frequency of each type of debug...
Similar to #17 which is about a more fine-grained performance benchmark for the rpcs, I suggest to do the same for internal functions such as `ConnectBlock`. We could either parse...
We already benchmark IBD (connecting blocks), so we might as well add another step after that to disconnect the blocks again with invalidateblock
@sdaftuar mentioned that there exists some tool to benchmark mining performance. See https://botbot.me/freenode/bitcoin-core-dev/2018-07-20/?msg=102302481&page=3 We should evaluate if it can be run by the performance monitor on a regular basis.