msg/async/rdma: fix a potential coredump when handling tx_buffers under heavy RDMA

[ownedu]

traffic, there are chances to access a current_chunk which can be beyond the
range of pre-allocated Tx buffer pool thus causes a coredump.

Signed-off-by: Yan Lei yongyou.yl@alibaba-inc.com

[ownedu]

@yuyuyu101 Pls help take a look at this PR; thanks!

[ownedu]

In case of OSD backfill, some OSD processes could crash with the following stack:

 1: (()+0x90da02) [0x7f4aa3403a02]
 2: (()+0xf100) [0x7f4aa1471100]
 3: (RDMAConnectedSocketImpl::submit(bool)+0xa3b) [0x7f4aa3638d7b]
 4: (RDMAWorker::handle_pending_message()+0xce) [0x7f4aa362608e]
 5: (EventCenter::process_events(int, std::chrono::duration<unsigned long, std::ratio<1l, 1000000000l> >*)+0x87c) [0x7f4aa35bbf4c]
 6: (()+0xb07ade) [0x7f4aa35fdade]
 7: (()+0xb5220) [0x7f4aa0174220]
 8: (()+0x7dc5) [0x7f4aa1469dc5]
 9: (clone()+0x6d) [0x7f4a9f8dbced]
 NOTE: a copy of the executable, or objdump -rdS <executable> is needed to interpret this.

[tchaikov]

@yuyuyu101 and @ownedu please avoid introducing merge commits.

[yuyuyu101]

my bad, I don't notice extra merge commit...

[ownedu]

Thanks @yuyuyu101

[ownedu]

@xiexingguo Could you pls add label RDMA for this PR? Thanks!