From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Sagi Grimberg <sagig@mellanox.com>,
Or Gerlitz <ogerlitz@mellanox.com>,
Nicholas Bellinger <nab@linux-iscsi.org>
Subject: [PATCH 3.10 85/85] iser-target: Fix post_send_buf_count for RDMA READ/WRITE
Date: Thu, 20 Mar 2014 17:10:28 -0700 [thread overview]
Message-ID: <20140321000610.906780970@linuxfoundation.org> (raw)
In-Reply-To: <20140321000558.606667505@linuxfoundation.org>
3.10-stable review patch. If anyone has any objections, please let me know.
------------------
From: Nicholas Bellinger <nab@linux-iscsi.org>
commit b6b87a1df604678ed1be40158080db012a99ccca upstream.
This patch fixes the incorrect setting of ->post_send_buf_count
related to RDMA WRITEs + READs where isert_rdma_rw->send_wr_num
was not being taken into account.
This includes incrementing ->post_send_buf_count within
isert_put_datain() + isert_get_dataout(), decrementing within
__isert_send_completion() + isert_response_completion(), and
clearing wr->send_wr_num within isert_completion_rdma_read()
This is necessary because even though IB_SEND_SIGNALED is
not set for RDMA WRITEs + READs, during a QP failure event
the work requests will be returned with exception status
from the TX completion queue.
Acked-by: Sagi Grimberg <sagig@mellanox.com>
Cc: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Nicholas Bellinger <nab@linux-iscsi.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/infiniband/ulp/isert/ib_isert.c | 14 ++++++++------
1 file changed, 8 insertions(+), 6 deletions(-)
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -1308,6 +1308,7 @@ isert_completion_rdma_read(struct iser_t
}
cmd->write_data_done = se_cmd->data_length;
+ wr->send_wr_num = 0;
pr_debug("isert_do_rdma_read_comp, calling target_execute_cmd\n");
spin_lock_bh(&cmd->istate_lock);
@@ -1367,6 +1368,7 @@ isert_response_completion(struct iser_tx
struct ib_device *ib_dev)
{
struct iscsi_cmd *cmd = &isert_cmd->iscsi_cmd;
+ struct isert_rdma_wr *wr = &isert_cmd->rdma_wr;
if (cmd->i_state == ISTATE_SEND_TASKMGTRSP ||
cmd->i_state == ISTATE_SEND_LOGOUTRSP ||
@@ -1377,7 +1379,7 @@ isert_response_completion(struct iser_tx
queue_work(isert_comp_wq, &isert_cmd->comp_work);
return;
}
- atomic_dec(&isert_conn->post_send_buf_count);
+ atomic_sub(wr->send_wr_num + 1, &isert_conn->post_send_buf_count);
cmd->i_state = ISTATE_SENT_STATUS;
isert_completion_put(tx_desc, isert_cmd, ib_dev);
@@ -1415,7 +1417,7 @@ isert_send_completion(struct iser_tx_des
case ISER_IB_RDMA_READ:
pr_debug("isert_send_completion: Got ISER_IB_RDMA_READ:\n");
- atomic_dec(&isert_conn->post_send_buf_count);
+ atomic_sub(wr->send_wr_num, &isert_conn->post_send_buf_count);
isert_completion_rdma_read(tx_desc, isert_cmd);
break;
default:
@@ -1834,12 +1836,12 @@ isert_put_datain(struct iscsi_conn *conn
isert_init_tx_hdrs(isert_conn, &isert_cmd->tx_desc);
isert_init_send_wr(isert_cmd, &isert_cmd->tx_desc.send_wr);
- atomic_inc(&isert_conn->post_send_buf_count);
+ atomic_add(wr->send_wr_num + 1, &isert_conn->post_send_buf_count);
rc = ib_post_send(isert_conn->conn_qp, wr->send_wr, &wr_failed);
if (rc) {
pr_warn("ib_post_send() failed for IB_WR_RDMA_WRITE\n");
- atomic_dec(&isert_conn->post_send_buf_count);
+ atomic_sub(wr->send_wr_num + 1, &isert_conn->post_send_buf_count);
}
pr_debug("Posted RDMA_WRITE + Response for iSER Data READ\n");
return 1;
@@ -1942,12 +1944,12 @@ isert_get_dataout(struct iscsi_conn *con
data_left -= data_len;
}
- atomic_inc(&isert_conn->post_send_buf_count);
+ atomic_add(wr->send_wr_num, &isert_conn->post_send_buf_count);
rc = ib_post_send(isert_conn->conn_qp, wr->send_wr, &wr_failed);
if (rc) {
pr_warn("ib_post_send() failed for IB_WR_RDMA_READ\n");
- atomic_dec(&isert_conn->post_send_buf_count);
+ atomic_sub(wr->send_wr_num, &isert_conn->post_send_buf_count);
}
pr_debug("Posted RDMA_READ memory for ISER Data WRITE\n");
return 0;
next prev parent reply other threads:[~2014-03-21 0:10 UTC|newest]
Thread overview: 90+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-03-21 0:09 [PATCH 3.10 00/85] 3.10.34-stable review Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 01/85] ocfs2: fix quota file corruption Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 02/85] ocfs2 syncs the wrong range Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 03/85] sched: Fix double normalization of vruntime Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 04/85] rapidio/tsi721: fix tasklet termination in dma channel release Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 05/85] net-tcp: fastopen: fix high order allocations Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 06/85] neigh: recompute reachabletime before returning from neigh_periodic_work() Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 07/85] virtio-net: alloc big buffers also when guest can receive UFO Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 08/85] ipv6: reuse ip6_frag_id from ip6_ufo_append_data Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 09/85] sfc: check for NULL efx->ptp_data in efx_ptp_event Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 10/85] ipv6: ipv6_find_hdr restore prev functionality Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 11/85] tg3: Dont check undefined error bits in RXBD Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 12/85] net: sctp: fix sctp_sf_do_5_1D_ce to verify if we/peer is AUTH capable Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 13/85] mac80211: send control port protocol frames to the VO queue Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 14/85] mac80211: fix AP powersave TX vs. wakeup race Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 15/85] mac80211: dont validate unchanged AP bandwidth while tracking Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 16/85] mac80211: fix association to 20/40 MHz VHT networks Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 17/85] mac80211: clear sequence/fragment number in QoS-null frames Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 18/85] ath9k: Fix ETSI compliance for AR9462 2.0 Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 19/85] iwlwifi: dvm: clear IWL_STA_UCODE_INPROGRESS when assoc fails Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 20/85] iwlwifi: fix TX status for aggregated packets Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 21/85] iwlwifi: disable TX AMPDU by default for iwldvm Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 22/85] mwifiex: clean pcie ring only when device is present Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 23/85] mwifiex: add NULL check for PCIe Rx skb Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 24/85] mwifiex: fix cmd and Tx data timeout issue for PCIe cards Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 25/85] mwifiex: do not advertise usb autosuspend support Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 26/85] mwifiex: copy APs HT capability info correctly Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 27/85] mwifiex: save and copy APs VHT " Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 28/85] ARM: 7811/1: locks: use early clobber in arch_spin_trylock Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 30/85] ALSA: oxygen: Xonar DG(X): capture from I2S channel 1, not 2 Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 31/85] ALSA: usb-audio: Add quirk for Logitech Webcam C500 Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 32/85] ALSA: hda - Added inverted digital-mic handling for Acer TravelMate 8371 Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 33/85] ALSA: hda - Add missing loopback merge path for AD1884/1984 codecs Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 34/85] powerpc: Align p_dyn, p_rela and p_st symbols Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 35/85] ARM: 7991/1: sa1100: fix compile problem on Collie Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 36/85] regulator: core: Replace direct ops->enable usage Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 37/85] x86: Ignore NMIs that come in during early boot Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 38/85] x86: fix compile error due to X86_TRAP_NMI use in asm files Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 39/85] x86/amd/numa: Fix northbridge quirk to assign correct NUMA node Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 40/85] usb: Add device quirk for Logitech HD Pro Webcams C920 and C930e Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 41/85] usb: Make DELAY_INIT quirk wait 100ms between Get Configuration requests Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 42/85] genirq: Remove racy waitqueue_active check Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 43/85] cpuset: fix a race condition in __cpuset_node_allowed_softwall() Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 44/85] ACPI / resources: ignore invalid ACPI device resources Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 45/85] tracing: Do not add event files for modules that fail tracepoints Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 46/85] firewire: net: fix use after free Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 47/85] firewire: dont use PREPARE_DELAYED_WORK Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 48/85] libata: add ATA_HORKAGE_BROKEN_FPDMA_AA quirk for Seagate Momentus SpinPoint M8 (2BA30001) Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 49/85] spi: spi-ath79: fix initial GPIO CS line setup Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 50/85] NFS: Fix a delegation callback race Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 51/85] NFSv4: nfs4_stateid_is_current should return true for an invalid stateid Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 52/85] ACPI / sleep: Add extra checks for HW Reduced ACPI mode sleep states Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 53/85] iscsi-target: Fix iscsit_get_tpg_from_np tpg_state bug Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 54/85] fs/proc/base.c: fix GPF in /proc/$PID/map_files Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 55/85] drm/radeon/atom: select the proper number of lanes in transmitter setup Greg Kroah-Hartman
2014-03-21 0:09 ` [PATCH 3.10 56/85] ASoC: pcm: free path list before exiting from error conditions Greg Kroah-Hartman
2014-03-22 18:55 ` Mark Brown
2014-03-24 4:34 ` Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 57/85] ipc: Fix 2 bugs in msgrcv() MSG_COPY implementation Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 59/85] PCI: Enable INTx in pci_reenable_device() only when MSI/MSI-X not enabled Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 60/85] vmxnet3: fix netpoll race condition Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 61/85] vmxnet3: fix building without CONFIG_PCI_MSI Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 62/85] mm/compaction: break out of loop on !PageBuddy in isolate_freepages_block Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 63/85] dm cache: fix truncation bug when copying a block to/from >2TB fast device Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 64/85] dm cache: fix access beyond end of origin device Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 65/85] net: unix socket code abuses csum_partial Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 66/85] can: flexcan: flexcan_open(): fix error path if flexcan_chip_start() fails Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 67/85] SCSI: isci: fix reset timeout handling Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 68/85] SCSI: isci: correct erroneous for_each_isci_host macro Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 69/85] SCSI: qla2xxx: Poll during initialization for ISP25xx and ISP83xx Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 70/85] SCSI: storvsc: NULL pointer dereference fix Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 71/85] x86, fpu: Check tsk_used_math() in kernel_fpu_end() for eager FPU Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 72/85] Btrfs: fix data corruption when reading/updating compressed extents Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 73/85] ALSA: oxygen: modify adjust_dg_dac_routing function Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 74/85] jiffies: Avoid undefined behavior from signed overflow Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 75/85] s390/dasd: hold request queue sysfs lock when calling elevator_init() Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 76/85] Fix mountpoint reference leakage in linkat Greg Kroah-Hartman
2014-03-21 0:11 ` Oleg Drokin
2014-03-21 0:38 ` Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 78/85] bio-integrity: Fix bio_integrity_verify segment start bug Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 79/85] tick: Make oneshot broadcast robust vs. CPU offlining Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 80/85] iwlwifi: mvm: dont WARN when statistics are handled late Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 81/85] ARM: 7864/1: Handle 64-bit memory in case of 32-bit phys_addr_t Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 82/85] ARM: ignore memory below PHYS_OFFSET Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 83/85] iscsi/iser-target: Use list_del_init for ->i_conn_node Greg Kroah-Hartman
2014-03-21 0:10 ` [PATCH 3.10 84/85] iscsi/iser-target: Fix isert_conn->state hung shutdown issues Greg Kroah-Hartman
2014-03-21 0:10 ` Greg Kroah-Hartman [this message]
2014-03-21 5:28 ` [PATCH 3.10 00/85] 3.10.34-stable review Guenter Roeck
2014-03-22 21:56 ` Shuah Khan
2014-03-24 4:35 ` Greg Kroah-Hartman
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20140321000610.906780970@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=linux-kernel@vger.kernel.org \
--cc=nab@linux-iscsi.org \
--cc=ogerlitz@mellanox.com \
--cc=sagig@mellanox.com \
--cc=stable@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).