rpma
rpma copied to clipboard
pmem
Remote Persistent Memory Access Library
Move on to the new coverage uploader. The old bash uploader is long EOL now. See: https://github.com/pmem/pmdk/pull/5548 - - - This change is [](https://reviewable.io/reviews/pmem/rpma/2126)
@grom72 @ldorau Do you plan to implement rpma verify based on traditional RDMA API? If so, we can talk about the design of rpma verify here. Best Regards, Xiao Yang
Figure 10. Bandwidth (threads): rpma_read() from DRAM (non-DW2PMem) vs from PMEM (DW2PMem) shall be extended by PMem (non-DW2PMem) series
# FEAT: RPMA Fio Engine (server-side) support for ``offset`` ## Rationale To enable running several engines sharing the same pmem file an offset is needed as a supported command option....
It might be a good idea to track if building Python documentation at least does not produce any errors. It can be easily added to the current build system e.g....
Add a superclass (abstract) Runner with all common code - see: #1427
It often happens: ``` [thread #0] mtt.c:80 mtt_thread_main() -> pthread_cond_timedwait() failed: Connection timed out ``` https://app.circleci.com/pipelines/github/ldorau/rpma/4534/workflows/8355d44e-07fd-4be6-b68e-fd15b5db05b4/jobs/4626
Some MT tests run under valgrind's memcheck sometimes hangs: https://app.circleci.com/pipelines/github/haichangsi/rpma/120/workflows/b2f790b5-0bd9-4852-8ac9-1dd45dac2578/jobs/120
# FEAT: rpma_cq_wait() performance optimization ## Rationale ``ibv_ack_cq_events()`` seems to be the main bottleneck for the librpma library when the completion event channel is used. To avoid the problem ``ibv_ack_cq_events()``...
# FEAT: rpma_conn_wait() performance optimization ## Rationale Having ``rpma_conn_wait()`` a part of the librpma API we can implement a similar solution as the one described in #1697 but also give...