On Sun, 2014-12-07 at 13:11 +0200, Sagi Grimberg wrote: > Hey Nic, > > So I figured it would be easier to just resend the series with the small fixup > instead of sending the specific bits and generating you some extra overhead. > > Changes from v3: > - Fixed the logging changes wrong squash (patch 32). > - Removed unneeded (at this point) patch: "Don't clear transport private cmd context" > - Added a warning fix by nab > > Nicholas Bellinger (1): > iser-target: Fix wc->wr_id cast warning > > Sagi Grimberg (34): > iscsi-target: Add call to wait_conn in establishment error flow > iser-target: Destroy the connection when getting a connect error > event > iser-target: Initiate connection termination only once > iser-target: Don't deffer disconnected handler to a work > iser-target: Reject connect request in failure path > iser-target: Introduce ISER_CONN_FULL_FEATURE state > iser-target: Parallelize CM connection establishment > iser-target: Use kref_get_unless_zero in connected_handler > iser-target: Acquire conn_mutex when changing connection state > iser-target: Don't override cm_id context > iser-target: Move cma_id setup to a function > iser-target: Handle ADDR_CHANGE event for listener cm_id > iser-target: Deffer the last part of wait_conn to a work > iser-target: Decrement completion context active_qps in error flow > iser-target: Move PI context allocation to a function > iser-target: Move PI handling to a routine > iser-target: Allocate PI contexts dynamically > iser-target: Fix NULL dereference in SW mode DIF > iscsi,iser-target: Expose supported protection ops according to > t10_pi > iser-target: Work-around live target stack shutdown resource cleanup > iser-target: Remove interrupt coalescing > iser-target: Unite error completion handler for RX and TX > iser-target: Cast wr_id with uintptr_t instead of unsinged long > iser-target: Centralize completion elements to a context > iser-target: Use single CQ for TX and RX > iser-target: Remove redundant call to isert_conn_terminate > iser-target: Remove an atomic operation from the IO path > iser-target: Introduce isert_poll_budget > iser-target: Reduce CQ lock contention by batch polling > iser-target: Don't wait for session commands from completion context > iser-target: Fix logout sequence > iser-target: Use debug_level parameter to control logging level > iser-target: Adjust log levels and prettify some prints > iser-target: Remove code duplication > > drivers/infiniband/ulp/isert/ib_isert.c | 1599 +++++++++++++++-------------- > drivers/infiniband/ulp/isert/ib_isert.h | 80 ++- > drivers/target/iscsi/iscsi_target_login.c | 10 +- > 3 files changed, 893 insertions(+), 796 deletions(-) > Ok, I finally went through and squashed a number of patches between 01-19 that where requested for stable in order to get something that resembles a stable patch series.. The end result for v3.10.y is: iscsi,iser-target: Initiate termination only once iser-target: Fix flush + disconnect completion handling iser-target: Parallelize CM connection establishment iser-target: Fix connected_handler + teardown flow race iser-target: Handle ADDR_CHANGE event for listener cm_id iser-target: Fix implicit termination of connections drivers/infiniband/ulp/isert/ib_isert.c | 344 +++++++++++++++++++++++++++++++++++--------------- drivers/infiniband/ulp/isert/ib_isert.h | 7 +- drivers/target/iscsi/iscsi_target_login.c | 3 + 3 files changed, 250 insertions(+), 104 deletions(-) and for v3.14.y: iser-target: Allocate PI contexts dynamically iser-target: Fix NULL dereference in SW mode DIF iscsi,iser-target: Expose supported protection ops according to t10_pi drivers/infiniband/ulp/isert/ib_isert.c | 298 +++++++++++++++++++++++++++++--------------------- drivers/infiniband/ulp/isert/ib_isert.h | 8 ++ drivers/target/iscsi/iscsi_target_login.c | 7 +- 3 files changed, 186 insertions(+), 127 deletions(-) For the moment the details about which patches are squashed into where is still in the commit logs in target-pending/for-next, but these will be dropped before the PULL request goes out next week. Beyond that, the only extra code change that was made was to go ahead and drop the now unused iscsi_np->tpg_np pointer. Please have a look in for-next and let me know if anything doesn't look right. Thank you, --nab -- To unsubscribe from this list: send the line "unsubscribe target-devel" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
