The following series implements...
1. fcoe: This patch simply implements round-robin based selection of CPU
for post-processing of command (for FCoE target)
2. fcoe: This patch amends first patch, it excludes the CPU which recived the
incoming request for post-processing and select other CPU. It also
addresses problem where, once rx_id is assigned, same CPU (CPU which
allocated the exchange from its per CPU pool) shall do post-processing
of incoming request including releasing exchange to the respective
per-CPU pool of exchanges.
This series contains amendment to "CPU selection patch" based on Andy Grover's
comment.
Patches which were amended:
Patch 1 -- fcoe: This patch simply implements round-robin based selection of
CPU
for post-processing of command (for FCoE target). Comment "to use cpumask*
functions"
is addressed. As well enhanced the patch to take care of UP system, CPU
going offline
after it was selected as a CPU to handle post-processing of command.
Dependencies:
---
Kiran Patil (2):
fcoe: Amends previous patch, Round-robin based selection of CPU for post
processing of incoming request for FCoE target
fcoe: Round-robin based selection of CPU for post-processing of incoming
commands
drivers/scsi/fcoe/fcoe.c | 43 ++++++++++++++++++++++++++++++++++++++++++-
1 files changed, 42 insertions(+), 1 deletions(-)
--
Thanks,
-- Kiran P
_______________________________________________
devel mailing list
[email protected]
https://lists.open-fcoe.org/mailman/listinfo/devel
