[PATCHv4 net-next 21/32] iw_cxgb4: adjust tcp snd/rcv window based on link speed

From: Steve Wise sw...@opengridcomputing.com 40G devices need a bigger windows, so default 40G devices to snd 512K rcv 1024K. Fixed a bug that shows up with recv window sizes that exceed the size of the RCV_BUFSIZ field in opt0 (= 1024K :). If the recv window exceeds this, then we specify the

[PATCHv4 net-next 17/32] iw_cxgb4: don't leak skb in c4iw_uld_rx_handler()

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index

[PATCHv4 net-next 25/32] iw_cxgb4: rx_data() needs to hold the ep mutex

From: Steve Wise sw...@opengridcomputing.com To avoid racing with other threads doing close/flush/whatever, rx_data() should hold the endpoint mutex. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 16 +--- 1 file changed, 9

[PATCHv4 net-next 11/32] cxgb4/iw_cxgb4: Doorbell Drop Avoidance Bug Fixes

From: Steve Wise sw...@opengridcomputing.com The current logic suffers from a slow response time to disable user DB usage, and also fails to avoid DB FIFO drops under heavy load. This commit fixes these deficiencies and makes the avoidance logic more optimal. This is done by more efficiently

[PATCHv4 net-next 15/32] iw_cxgb4: default peer2peer mode to 1

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCHv4 net-next 32/32] iw_cxgb4: Use pr_warn_ratelimited

Signed-off-by: Hariprasad Shenai haripra...@chelsio.com --- drivers/infiniband/hw/cxgb4/resource.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/resource.c b/drivers/infiniband/hw/cxgb4/resource.c index d9bc9ba..67df71a 100644 --- a/drivers

[PATCHv4 net-next 22/32] iw_cxgb4: update snd_seq when sending MPA messages

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index f6891b8..95b3c01 100644 ---

[PATCHv4 net-next 26/32] iw_cxgb4: endpoint timeout fixes

From: Steve Wise sw...@opengridcomputing.com 1) timedout endpoint processing can be starved. If there is continual CPL messages flowing into the driver, the endpoint timeout processing can be starved. This condition exposed the other bugs below. Solution: In process_work(), call

[PATCHv4 net-next 30/32] iw_cxgb4: minor fixes

From: Steve Wise sw...@opengridcomputing.com Added some missing debug stats. Use uninitialized_var(). Initialize reserved fields in a FW work request. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 2 +- drivers/infiniband/hw/cxgb4/mem.c

[PATCHv4 net-next 29/32] iw_cxgb4: SQ flush fix

From: Steve Wise sw...@opengridcomputing.com There is a race when moving a QP from RTS-CLOSING where a SQ work request could be posted after the FW receives the RDMA_RI/FINI WR. The SQ work request will never get processed, and should be completed with FLUSHED status. Function c4iw_flush_sq(),

[PATCHv4 net-next 13/32] iw_cxgb4: Fix incorrect BUG_ON conditions

From: Steve Wise sw...@opengridcomputing.com Based on original work from Jay Hernandez j...@chelsio.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

[PATCHv4 net-next 08/32] iw_cxgb4: Allow loopback connections

From: Steve Wise sw...@opengridcomputing.com find_route() must treat loopback as a valid egress interface. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git

[PATCHv4 net-next 20/32] iw_cxgb4: connect_request_upcall fixes

From: Steve Wise sw...@opengridcomputing.com When processing an MPA Start Request, if the listening endpoint is DEAD, then abort the connection. If the IWCM returns an error, then we must abort the connection and release resources. Also abort_connection() should not post a CLOSE event, so clean

[PATCHv4 net-next 14/32] iw_cxgb4: Mind the sq_sig_all/sq_sig_type QP attributes

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 1 + drivers/infiniband/hw/cxgb4/qp.c | 6 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git

[PATCHv4 net-next 19/32] iw_cxgb4: ignore read reponse type 1 CQEs

From: Steve Wise sw...@opengridcomputing.com These are generated by HW in some error cases and need to be silently discarded. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 24 1 file changed, 20 insertions(+), 4

[PATCHv4 net-next 16/32] iw_cxgb4: save the correct map length for fast_reg_page_lists

From: Steve Wise sw...@opengridcomputing.com We cannot save the mapped length using the rdma max_page_list_len field of the ib_fast_reg_page_list struct because the core code uses it. This results in an incorrect unmap of the page list in c4iw_free_fastreg_pbl(). I found this with dma map

[PATCHv3 net-next 02/31] cxgb4: Add code to dump SGE registers when hitting idma hangs

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 1 + drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 106 +++

[PATCHv3 net-next 01/31] cxgb4: Fix some small bugs in t4_sge_init_soft() when our Page Size is 64KB

From: Kumar Sanghvi kuma...@chelsio.com We'd come in with SGE_FL_BUFFER_SIZE[0] and [1] both equal to 64KB and the extant logic would flag that as an error. Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com ---

[PATCHv3 net-next 03/31] cxgb4: Rectify emitting messages about SGE Ingress DMA channels being potentially stuck

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h | 9 ++- drivers/net/ethernet/chelsio/cxgb4/sge.c | 91 -- 2 files

[PATCHv3 net-next 27/31] iw_cxgb4: wc_wmb() needed after DB writes.

From: Steve Wise sw...@opengridcomputing.com Need to do an sfence after both the WC and regular PIDX DB write. Otherwise the host might reorder things and cause work request corruption (seen with NFSRDMA). Signed-off-by: Steve Wise sw...@opengridcomputing.com ---

[PATCHv3 net-next 28/31] iw_cxgb4: SQ flush fix.

From: Steve Wise sw...@opengridcomputing.com There is a race when moving a QP from RTS-CLOSING where a SQ work request could be posted after the FW receives the RDMA_RI/FINI WR. The SQ work request will never get processed, and should be completed with FLUSHED status. Function c4iw_flush_sq(),

[PATCHv3 net-next 09/31] iw_cxgb4: Treat CPL_ERR_KEEPALV_NEG_ADVICE as negative advice.

From: Steve Wise sw...@opengridcomputing.com Based on original work by Anand Priyadarshee ana...@chelsio.com. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c| 24 drivers/net/ethernet/chelsio/cxgb4/t4_msg.h | 1 +

[PATCHv3 net-next 26/31] iw_cxgb4: rmb() after reading valid gen bit.

From: Steve Wise sw...@opengridcomputing.com Some HW platforms can reorder read operations, so we must rmb() after we see a valid gen bit in a CQE but before we read any other fields from the CQE. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/t4.h | 1 +

[PATCHv3 net-next 25/31] iw_cxgb4: endpoint timeout fixes.

From: Steve Wise sw...@opengridcomputing.com 1) timedout endpoint processing can be starved. If there is continual CPL messages flowing into the driver, the endpoint timeout processing can be starved. This condition exposed the other bugs below. Solution: In process_work(), call

[PATCHv3 net-next 13/31] iw_cxgb4: Mind the sq_sig_all/sq_sig_type QP attributes.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 1 + drivers/infiniband/hw/cxgb4/qp.c | 6 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git

[PATCHv3 net-next 07/31] iw_cxgb4: Allow loopback connections.

From: Steve Wise sw...@opengridcomputing.com find_route() must treat loopback as a valid egress interface. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) diff --git

[PATCHv3 net-next 20/31] iw_cxgb4: adjust tcp snd/rcv window based on link speed.

From: Steve Wise sw...@opengridcomputing.com 40G devices need a bigger windows, so default 40G devices to snd 512K rcv 1024K. Fixed a bug that shows up with recv window sizes that exceed the size of the RCV_BUFSIZ field in opt0 (= 1024K :). If the recv window exceeds this, then we specify the

[PATCHv3 net-next 11/31] iw_cxgb4: use the BAR2/WC path for kernel QPs and T5 devices.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c | 41 +- drivers/infiniband/hw/cxgb4/iw_cxgb4.h | 2 ++ drivers/infiniband/hw/cxgb4/qp.c | 59 +---

[PATCHv3 net-next 16/31] iw_cxgb4: don't leak skb in c4iw_uld_rx_handler().

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c | 6 -- 1 file changed, 4 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c index

[PATCHv3 net-next 21/31] iw_cxgb4: update snd_seq when sending MPA messages.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index f6891b8..95b3c01 100644 ---

[PATCHv3 net-next 12/31] iw_cxgb4: Fix incorrect BUG_ON conditions.

From: Steve Wise sw...@opengridcomputing.com Based on original work from Jay Hernandez j...@chelsio.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git

[PATCHv3 net-next 22/31] iw_cxgb4: lock around accept/reject downcalls.

From: Steve Wise sw...@opengridcomputing.com There is a race between ULP threads doing an accept/reject, and the ingress processing thread handling close/abort for the same connection. The accept/reject path needs to hold the lock to serialize these paths. Signed-off-by: Steve Wise

[PATCHv3 net-next 05/31] cxgb4: use spinlock_irqsave/spinlock_irqrestore for db lock.

From: Kumar Sanghvi kuma...@chelsio.com Currently ring_tx_db() can deadlock if a db_full interrupt fires and is run on the same while ring_tx_db() has the db lock held. It needs to disable interrupts since it serializes with an interrupt handler. Based on original work by Steve Wise

[PATCHv3 net-next 08/31] iw_cxgb4: release neigh entry in error paths.

From: Steve Wise sw...@opengridcomputing.com Always release the neigh entry in rx_pkt(). Based on original work by Santosh Rastapur sant...@chelsio.com. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 5 +++-- 1 file changed, 3 insertions(+), 2

[PATCHv3 net-next 14/31] iw_cxgb4: default peer2peer mode to 1.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCHv3 net-next 24/31] iw_cxgb4: rx_data() needs to hold the ep mutex.

From: Steve Wise sw...@opengridcomputing.com To avoid racing with other threads doing close/flush/whatever, rx_data() should hold the endpoint mutex. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 16 +--- 1 file changed, 9

[PATCHv3 net-next 23/31] iw_cxgb4: drop RX_DATA packets if the endpoint is gone.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index f4a9ebe..e7c6dfd 100644 ---

[PATCHv3 net-next 31/31] Revert cxgb4: Don't assume LSO only uses SGL path in t4_eth_xmit()

Commit 0034b29 (cxgb4: Don't assume LSO only uses SGL path in t4_eth_xmit()) introduced a regression causing chip-hang. This patch needs more debugging and more work. So reverting for now. Signed-off-by: Hariprasad Shenai haripra...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/sge.c | 32

[PATCHv3 net-next 29/31] iw_cxgb4: minor fixes

From: Steve Wise sw...@opengridcomputing.com Added some missing debug stats. Use uninitialized_var(). Initialize reserved fields in a FW work request. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 2 +- drivers/infiniband/hw/cxgb4/mem.c

[PATCHv3 net-next 19/31] iw_cxgb4: connect_request_upcall fixes.

From: Steve Wise sw...@opengridcomputing.com When processing an MPA Start Request, if the listening endpoint is DEAD, then abort the connection. If the IWCM returns an error, then we must abort the connection and release resources. Also abort_connection() should not post a CLOSE event, so clean

[PATCHv3 net-next 04/31] cxgb4: Updates for T5 SGE's Egress Congestion Threshold

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/sge.c | 18 +- drivers/net/ethernet/chelsio/cxgb4/t4_regs.h | 6 ++ 2 files changed,

[PATCHv3 net-next 06/31] iw_cxgb4: cap CQ size at T4_MAX_IQ_SIZE.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index 88de3aa..c0673ac

[PATCHv3 net-next 10/31] cxgb4/iw_cxgb4: Doorbell Drop Avoidance Bug Fixes.

From: Steve Wise sw...@opengridcomputing.com The current logic suffers from a slow response time to disable user DB usage, and also fails to avoid DB FIFO drops under heavy load. This commit fixes these deficiencies and makes the avoidance logic more optimal. This is done by more efficiently

[PATCHv3 net-next 18/31] iw_cxgb4: ignore read reponse type 1 CQEs.

From: Steve Wise sw...@opengridcomputing.com These are generated by HW in some error cases and need to be silently discarded. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 24 1 file changed, 20 insertions(+), 4

[PATCHv3 net-next 30/31] iw_cxgb4: Max fastreg depth depends on DSGL support.

From: Steve Wise sw...@opengridcomputing.com The max depth of a fastreg mr depends on whether the device supports DSGL or not. So compute it dynamically based on the device support and the module use_dsgl option. Signed-off-by: Steve Wise sw...@opengridcomputing.com ---

[PATCHv3 net-next 17/31] iw_cxgb4: fix possible memory leak in RX_PKT processing.

From: Steve Wise sw...@opengridcomputing.com If cxgb4_ofld_send() returns 0, then send_fw_pass_open_req() must free the request skb and the saved skb with the tcp header. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 9 - 1 file changed, 8

[PATCHv2 net-next 00/31] Misc. fixes for cxgb4 and iw_cxgb4

. (cxgb4/iw_cxgb4: Doorbell Drop Avoidance Bug Fixes.) Hariprasad Shenai (1): Revert cxgb4: Don't assume LSO only uses SGL path in t4_eth_xmit() Kumar Sanghvi (5): cxgb4: Fix some small bugs in t4_sge_init_soft() when our Page Size is 64KB cxgb4: Add code to dump SGE registers when hitting

[PATCHv2 net-next 06/31] iw_cxgb4: cap CQ size at T4_MAX_IQ_SIZE.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index

[PATCHv2 net-next 02/31] cxgb4: Add code to dump SGE registers when hitting idma hangs

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h |1 + drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 106 ++

[PATCHv2 net-next 01/31] cxgb4: Fix some small bugs in t4_sge_init_soft() when our Page Size is 64KB

From: Kumar Sanghvi kuma...@chelsio.com We'd come in with SGE_FL_BUFFER_SIZE[0] and [1] both equal to 64KB and the extant logic would flag that as an error. Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com ---

[PATCHv2 net-next 04/31] cxgb4: Updates for T5 SGE's Egress Congestion Threshold

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/sge.c | 18 +- drivers/net/ethernet/chelsio/cxgb4/t4_regs.h |6 ++ 2 files

[PATCHv2 net-next 07/31] iw_cxgb4: Allow loopback connections.

From: Steve Wise sw...@opengridcomputing.com find_route() must treat loopback as a valid egress interface. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git

[PATCHv2 net-next 08/31] iw_cxgb4: release neigh entry in error paths.

From: Steve Wise sw...@opengridcomputing.com Always release the neigh entry in rx_pkt(). Based on original work by Santosh Rastapur sant...@chelsio.com. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |5 +++-- 1 files changed, 3 insertions(+), 2

[PATCHv2 net-next 09/31] iw_cxgb4: Treat CPL_ERR_KEEPALV_NEG_ADVICE as negative advice.

From: Steve Wise sw...@opengridcomputing.com Based on original work by Anand Priyadarshee ana...@chelsio.com. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c| 25 + drivers/net/ethernet/chelsio/cxgb4/t4_msg.h |

[PATCHv2 net-next 12/31] iw_cxgb4: Fix incorrect BUG_ON conditions.

From: Steve Wise sw...@opengridcomputing.com Based on original work from Jay Hernandez j...@chelsio.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git

[PATCHv2 net-next 30/31] iw_cxgb4: Max fastreg depth depends on DSGL support.

From: Steve Wise sw...@opengridcomputing.com The max depth of a fastreg mr depends on whether the device supports DSGL or not. So compute it dynamically based on the device support and the module use_dsgl option. Signed-off-by: Steve Wise sw...@opengridcomputing.com ---

[PATCHv2 net-next 22/31] iw_cxgb4: lock around accept/reject downcalls.

From: Steve Wise sw...@opengridcomputing.com There is a race between ULP threads doing an accept/reject, and the ingress processing thread handling close/abort for the same connection. The accept/reject path needs to hold the lock to serialize these paths. Signed-off-by: Steve Wise

[PATCHv2 net-next 29/31] iw_cxgb4: minor fixes/cleanup.

From: Steve Wise sw...@opengridcomputing.com Added some missing debug stats. Use uninitialized_var(). Rate limit warning printks. Initialize reserved fields in a FW work request. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c |2 +-

[PATCHv2 net-next 28/31] iw_cxgb4: SQ flush fix.

From: Steve Wise sw...@opengridcomputing.com There is a race when moving a QP from RTS-CLOSING where a SQ work request could be posted after the FW receives the RDMA_RI/FINI WR. The SQ work request will never get processed, and should be completed with FLUSHED status. Function c4iw_flush_sq(),

[PATCHv2 net-next 15/31] iw_cxgb4: save the correct map length for fast_reg_page_lists.

From: Steve Wise sw...@opengridcomputing.com We cannot save the mapped length using the rdma max_page_list_len field of the ib_fast_reg_page_list struct because the core code uses it. This results in an incorrect unmap of the page list in c4iw_free_fastreg_pbl(). I found this with dma map

[PATCHv2 net-next 16/31] iw_cxgb4: don't leak skb in c4iw_uld_rx_handler().

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c |4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c

[PATCHv2 net-next 10/31] cxgb4/iw_cxgb4: Doorbell Drop Avoidance Bug Fixes.

From: Steve Wise sw...@opengridcomputing.com The current logic suffers from a slow response time to disable user DB usage, and also fails to avoid DB FIFO drops under heavy load. This commit fixes these deficiencies and makes the avoidance logic more optimal. This is done by more efficiently

[PATCHv2 net-next 13/31] iw_cxgb4: Mind the sq_sig_all/sq_sig_type QP attributes.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/iw_cxgb4.h |1 + drivers/infiniband/hw/cxgb4/qp.c |6 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git

[PATCHv2 net-next 11/31] iw_cxgb4: use the BAR2/WC path for kernel QPs and T5 devices.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c | 41 + drivers/infiniband/hw/cxgb4/iw_cxgb4.h |2 + drivers/infiniband/hw/cxgb4/qp.c | 59

[PATCHv2 net-next 05/31] cxgb4: use spinlock_irqsave/spinlock_irqrestore for db lock.

From: Kumar Sanghvi kuma...@chelsio.com Currently ring_tx_db() can deadlock if a db_full interrupt fires and is run on the same while ring_tx_db() has the db lock held. It needs to disable interrupts since it serializes with an interrupt handler. Based on original work by Steve Wise

[PATCHv2 net-next 14/31] iw_cxgb4: default peer2peer mode to 1.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCHv2 net-next 17/31] iw_cxgb4: fix possible memory leak in RX_PKT processing.

From: Steve Wise sw...@opengridcomputing.com If cxgb4_ofld_send() returns 0, then send_fw_pass_open_req() must free the request skb and the saved skb with the tcp header. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |9 - 1 files

[PATCHv2 net-next 20/31] iw_cxgb4: adjust tcp snd/rcv window based on link speed.

From: Steve Wise sw...@opengridcomputing.com 40G devices need a bigger windows, so default 40G devices to snd 512K rcv 1024K. Fixed a bug that shows up with recv window sizes that exceed the size of the RCV_BUFSIZ field in opt0 (= 1024K :). If the recv window exceeds this, then we specify the

[PATCHv2 net-next 18/31] iw_cxgb4: ignore read reponse type 1 CQEs.

From: Steve Wise sw...@opengridcomputing.com These are generated by HW in some error cases and need to be silently discarded. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 20 1 files changed, 20 insertions(+), 0 deletions(-)

[PATCHv2 net-next 21/31] iw_cxgb4: update snd_seq when sending MPA messages.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCHv2 net-next 23/31] iw_cxgb4: drop RX_DATA packets if the endpoint is gone.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCHv2 net-next 26/31] iw_cxgb4: rmb() after reading valid gen bit.

From: Steve Wise sw...@opengridcomputing.com Some HW platforms can reorder read operations, so we must rmb() after we see a valid gen bit in a CQE but before we read any other fields from the CQE. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/t4.h |1

[PATCHv2 net-next 25/31] iw_cxgb4: endpoint timeout fixes.

From: Steve Wise sw...@opengridcomputing.com 1) timedout endpoint processing can be starved. If there is continual CPL messages flowing into the driver, the endpoint timeout processing can be starved. This condition exposed the other bugs below. Solution: In process_work(), call

[PATCHv2 net-next 27/31] iw_cxgb4: wc_wmb() needed after DB writes.

From: Steve Wise sw...@opengridcomputing.com Need to do an sfence after both the WC and regular PIDX DB write. Otherwise the host might reorder things and cause work request corruption (seen with NFSRDMA). Signed-off-by: Steve Wise sw...@opengridcomputing.com ---

[PATCH net-next 06/31] iw_cxgb4: cap CQ size at T4_MAX_IQ_SIZE.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c |2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c index

[PATCH net-next 23/31] iw_cxgb4: drop RX_DATA packets if the endpoint is gone.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCH net-next 16/31] iw_cxgb4: don't leak skb in c4iw_uld_rx_handler().

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c |4 +++- 1 files changed, 3 insertions(+), 1 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c

[PATCH net-next 07/31] iw_cxgb4: Allow loopback connections.

From: Steve Wise sw...@opengridcomputing.com find_route() must treat loopback as a valid egress interface. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |3 ++- 1 files changed, 2 insertions(+), 1 deletions(-) diff --git

[PATCH net-next 18/31] iw_cxgb4: ignore read reponse type 1 CQEs.

From: Steve Wise sw...@opengridcomputing.com These are generated by HW in some error cases and need to be silently discarded. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c | 20 1 files changed, 20 insertions(+), 0 deletions(-)

[PATCH net-next 29/31] iw_cxgb4: minor fixes/cleanup.

From: Steve Wise sw...@opengridcomputing.com Added some missing debug stats. Use uninitialized_var(). Rate limit warning printks. Initialize reserved fields in a FW work request. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c |2 +-

[PATCH net-next 21/31] iw_cxgb4: update snd_seq when sending MPA messages.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/drivers/infiniband/hw/cxgb4/cm.c b/drivers/infiniband/hw/cxgb4/cm.c index

[PATCH net-next 12/31] iw_cxgb4: Fix incorrect BUG_ON conditions.

From: Steve Wise sw...@opengridcomputing.com Based on original work from Jay Hernandez j...@chelsio.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cq.c |4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git

[PATCH net-next 09/31] iw_cxgb4: Treat CPL_ERR_KEEPALV_NEG_ADVICE as negative advice.

From: Steve Wise sw...@opengridcomputing.com Based on original work by Anand Priyadarshee ana...@chelsio.com. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c| 25 + drivers/net/ethernet/chelsio/cxgb4/t4_msg.h |

[PATCH net-next 28/31] iw_cxgb4: SQ flush fix.

From: Steve Wise sw...@opengridcomputing.com There is a race when moving a QP from RTS-CLOSING where a SQ work request could be posted after the FW receives the RDMA_RI/FINI WR. The SQ work request will never get processed, and should be completed with FLUSHED status. Function c4iw_flush_sq(),

[PATCH net-next 20/31] iw_cxgb4: adjust tcp snd/rcv window based on link speed.

From: Steve Wise sw...@opengridcomputing.com 40G devices need a bigger windows, so default 40G devices to snd 512K rcv 1024K. Fixed a bug that shows up with recv window sizes that exceed the size of the RCV_BUFSIZ field in opt0 (= 1024K :). If the recv window exceeds this, then we specify the

[PATCH net-next 26/31] iw_cxgb4: rmb() after reading valid gen bit.

From: Steve Wise sw...@opengridcomputing.com Some HW platforms can reorder read operations, so we must rmb() after we see a valid gen bit in a CQE but before we read any other fields from the CQE. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/t4.h |1

[PATCH net-next 25/31] iw_cxgb4: endpoint timeout fixes.

From: Steve Wise sw...@opengridcomputing.com 1) timedout endpoint processing can be starved. If there is continual CPL messages flowing into the driver, the endpoint timeout processing can be starved. This condition exposed the other bugs below. Solution: In process_work(), call

[PATCH net-next 02/31] cxgb4: Add code to dump SGE registers when hitting idma hangs

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/cxgb4.h |1 + drivers/net/ethernet/chelsio/cxgb4/t4_hw.c | 106 ++

[PATCH net-next 30/31] iw_cxgb4: Max fastreg depth depends on DSGL support.

From: Steve Wise sw...@opengridcomputing.com The max depth of a fastreg mr depends on whether the device supports DSGL or not. So compute it dynamically based on the device support and the module use_dsgl option. Signed-off-by: Steve Wise sw...@opengridcomputing.com ---

[PATCH net-next 05/31] cxgb4: use spinlock_irqsave/spinlock_irqrestore for db lock.

From: Kumar Sanghvi kuma...@chelsio.com Currently ring_tx_db() can deadlock if a db_full interrupt fires and is run on the same while ring_tx_db() has the db lock held. It needs to disable interrupts since it serializes with an interrupt handler. Based on original work by Steve Wise

[PATCH net-next 11/31] iw_cxgb4: use the BAR2/WC path for kernel QPs and T5 devices.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/device.c | 41 + drivers/infiniband/hw/cxgb4/iw_cxgb4.h |2 + drivers/infiniband/hw/cxgb4/qp.c | 59

[PATCH net-next 01/31] cxgb4: Fix some small bugs in t4_sge_init_soft() when our Page Size is 64KB

From: Kumar Sanghvi kuma...@chelsio.com We'd come in with SGE_FL_BUFFER_SIZE[0] and [1] both equal to 64KB and the extant logic would flag that as an error. Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com ---

[PATCH net-next 00/31] Misc. fixes for cxgb4 and iw_cxgb4

Hi All, This patch series provides miscelleneous fixes for Chelsio T4/T5 adapters related to cxgb4 related to sge and mtu. And includes DB Drop avoidance and other misc. fixes on iw-cxgb4. The patches series is created against David Miller's 'net-next' tree. And includes patches on cxgb4 and

[PATCH net-next 31/31] Revert cxgb4: Don't assume LSO only uses SGL path in t4_eth_xmit()

From: Kumar Sanghvi kuma...@chelsio.com Commit 0034b29 (cxgb4: Don't assume LSO only uses SGL path in t4_eth_xmit()) introduced a regression causing chip-hang. This patch needs more debugging and more work. So reverting for now. Signed-off-by: Hariprasad Shenai haripra...@chelsio.com

[PATCH net-next 24/31] iw_cxgb4: rx_data() needs to hold the ep mutex.

From: Steve Wise sw...@opengridcomputing.com To avoid racing with other threads doing close/flush/whatever, rx_data() should hold the endpoint mutex. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c | 16 +--- 1 files changed, 9

[PATCH net-next 04/31] cxgb4: Updates for T5 SGE's Egress Congestion Threshold

From: Kumar Sanghvi kuma...@chelsio.com Based on original work by Casey Leedom lee...@chelsio.com Signed-off-by: Kumar Sanghvi kuma...@chelsio.com --- drivers/net/ethernet/chelsio/cxgb4/sge.c | 18 +- drivers/net/ethernet/chelsio/cxgb4/t4_regs.h |6 ++ 2 files

[PATCH net-next 17/31] iw_cxgb4: fix possible memory leak in RX_PKT processing.

From: Steve Wise sw...@opengridcomputing.com If cxgb4_ofld_send() returns 0, then send_fw_pass_open_req() must free the request skb and the saved skb with the tcp header. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |9 - 1 files

[PATCH net-next 08/31] iw_cxgb4: release neigh entry in error paths.

From: Steve Wise sw...@opengridcomputing.com Always release the neigh entry in rx_pkt(). Based on original work by Santosh Rastapur sant...@chelsio.com. Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/cm.c |5 +++-- 1 files changed, 3 insertions(+), 2

[PATCH net-next 27/31] iw_cxgb4: wc_wmb() needed after DB writes.

From: Steve Wise sw...@opengridcomputing.com Need to do an sfence after both the WC and regular PIDX DB write. Otherwise the host might reorder things and cause work request corruption (seen with NFSRDMA). Signed-off-by: Steve Wise sw...@opengridcomputing.com ---

[PATCH net-next 13/31] iw_cxgb4: Mind the sq_sig_all/sq_sig_type QP attributes.

From: Steve Wise sw...@opengridcomputing.com Signed-off-by: Steve Wise sw...@opengridcomputing.com --- drivers/infiniband/hw/cxgb4/iw_cxgb4.h |1 + drivers/infiniband/hw/cxgb4/qp.c |6 -- 2 files changed, 5 insertions(+), 2 deletions(-) diff --git

<    1   2   3   4   >