xgboost survival trees in R: log-rank splitting and C-index

Having right-censored data (time to event, such as time to death) random survival forests in R using the R-function rfsrc with log-rank splitting and e.g. C-index for prediction error can be used.

I know xgboost has some options for survival data: cox regression and accelerated failure time. But that would not be the same as described above. How can I boost survival trees with log-rank splitting rule?