Hi Nic, here's a set of patches and fixes I've found during debugging FCoE over virtio. The first two fix a race condition I've encountered during starting and stopping initiators; they are similar to your prior patches, but I still think they should be applied to avoid any race issues. The next three are just minor cleanups, but the last one is _absolutely_ crucial. With the current code tcm_fc will behave _very_ odd if you're running in a virtualized environment, as the xid allocator in libfc will be thouroughly thrashed by sending frames on _other_ cpus than those the originator frame got received on. This causes frames to be delayed (by up to several minutes) leading to I/O errors on the initiator side.
As usual, comments and reviews are welcome. Hannes Reinecke (6): target: fix hang in target_wait_for_sess_cmds() target: fix potential race window in target_sess_cmd_list_waiting() target/tcm_fc: print command pointer in debug message target/tcm_fc: return detailed error in ft_sess_create() target/tcm_fc: Update debugging statements to match libfc usage target/tcm_fc: use CPU affinity for responses drivers/target/target_core_transport.c | 28 ++++++++++++++--------- drivers/target/tcm_fc/tfc_cmd.c | 4 ++-- drivers/target/tcm_fc/tfc_sess.c | 42 +++++++++++++++++++++++----------- 3 files changed, 48 insertions(+), 26 deletions(-) -- 1.8.5.6 -- To unsubscribe from this list: send the line "unsubscribe linux-scsi" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
