408728025e
The purpose of this patch is to fix the issue when there is no data buffer allocated, the previous method is wrong to set the recv pdu state. The reason is that: 1 When there is no data buffer allocated, we still need to handle the incoming pdu. It means that we should switch the pdu recv state immedidately. 2 And when there is a buffer, we resume the req handling with the allocated buffer, that time we should not switch the pdu receving state of the tqpair. Change-Id: I1cc2723acc7b0a17407c3a2e6273313a4e612916 Signed-off-by: Ziye Yang <optimistyzy@gmail.com> Reviewed-on: https://review.gerrithub.io/436153 Tested-by: SPDK CI Jenkins <sys_sgci@intel.com> Chandler-Test-Pool: SPDK Automated Test System <sys_sgsw@intel.com> Reviewed-by: Jim Harris <james.r.harris@intel.com> Reviewed-by: Ben Walker <benjamin.walker@intel.com> |
||
---|---|---|
.. | ||
ctrlr_bdev.c | ||
ctrlr_discovery.c | ||
ctrlr.c | ||
Makefile | ||
nvmf_fc.h | ||
nvmf_internal.h | ||
nvmf.c | ||
rdma.c | ||
request.c | ||
subsystem.c | ||
tcp.c | ||
transport.c | ||
transport.h |