From: Sasha Levin <Alexander.Levin@microsoft.com>
To: "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
"stable@vger.kernel.org" <stable@vger.kernel.org>
Cc: Bharat Potnuri <bharat@chelsio.com>,
Jason Gunthorpe <jgg@mellanox.com>,
Sasha Levin <Alexander.Levin@microsoft.com>
Subject: [PATCH AUTOSEL for 4.14 39/67] iser-target: avoid reinitializing rdma contexts for isert commands
Date: Thu, 8 Mar 2018 04:57:48 +0000 [thread overview]
Message-ID: <20180308045641.7814-39-alexander.levin@microsoft.com> (raw)
In-Reply-To: <20180308045641.7814-1-alexander.levin@microsoft.com>
From: Bharat Potnuri <bharat@chelsio.com>
[ Upstream commit 66f53e6f5400578bae58db0c06d85a8820831f40 ]
isert commands that failed during isert_rdma_rw_ctx_post() are queued to
Queue-Full(QF) queue and are scheduled to be reposted during queue-full
queue processing. During this reposting, the rdma contexts are initialised
again in isert_rdma_rw_ctx_post(), which is leaking significant memory.
unreferenced object 0xffff8830201d9640 (size 64):
comm "kworker/0:2", pid 195, jiffies 4295374851 (age 4528.436s)
hex dump (first 32 bytes):
00 60 8b cb 2e 00 00 00 00 10 00 00 00 00 00 00 .`..............
00 90 e3 cb 2e 00 00 00 00 10 00 00 00 00 00 00 ................
backtrace:
[<ffffffff8170711e>] kmemleak_alloc+0x4e/0xb0
[<ffffffff811f8ba5>] __kmalloc+0x125/0x2b0
[<ffffffffa046b24f>] rdma_rw_ctx_init+0x15f/0x6f0 [ib_core]
[<ffffffffa07ab644>] isert_rdma_rw_ctx_post+0xc4/0x3c0 [ib_isert]
[<ffffffffa07ad972>] isert_put_datain+0x112/0x1c0 [ib_isert]
[<ffffffffa07dddce>] lio_queue_data_in+0x2e/0x30 [iscsi_target_mod]
[<ffffffffa076c322>] target_qf_do_work+0x2b2/0x4b0 [target_core_mod]
[<ffffffff81080c3b>] process_one_work+0x1db/0x5d0
[<ffffffff8108107d>] worker_thread+0x4d/0x3e0
[<ffffffff81088667>] kthread+0x117/0x150
[<ffffffff81713fa7>] ret_from_fork+0x27/0x40
[<ffffffffffffffff>] 0xffffffffffffffff
Here is patch to use the older rdma contexts while reposting
the isert commands intead of reinitialising them.
Signed-off-by: Potnuri Bharat Teja <bharat@chelsio.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Sasha Levin <alexander.levin@microsoft.com>
---
drivers/infiniband/ulp/isert/ib_isert.c | 7 +++++++
drivers/infiniband/ulp/isert/ib_isert.h | 1 +
2 files changed, 8 insertions(+)
diff --git a/drivers/infiniband/ulp/isert/ib_isert.c b/drivers/infiniband/ulp/isert/ib_isert.c
index 9d4785ba24cb..e770c17cbca9 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.c
+++ b/drivers/infiniband/ulp/isert/ib_isert.c
@@ -2124,6 +2124,9 @@ isert_rdma_rw_ctx_post(struct isert_cmd *cmd, struct isert_conn *conn,
u32 rkey, offset;
int ret;
+ if (cmd->ctx_init_done)
+ goto rdma_ctx_post;
+
if (dir == DMA_FROM_DEVICE) {
addr = cmd->write_va;
rkey = cmd->write_stag;
@@ -2151,11 +2154,15 @@ isert_rdma_rw_ctx_post(struct isert_cmd *cmd, struct isert_conn *conn,
se_cmd->t_data_sg, se_cmd->t_data_nents,
offset, addr, rkey, dir);
}
+
if (ret < 0) {
isert_err("Cmd: %p failed to prepare RDMA res\n", cmd);
return ret;
}
+ cmd->ctx_init_done = true;
+
+rdma_ctx_post:
ret = rdma_rw_ctx_post(&cmd->rw, conn->qp, port_num, cqe, chain_wr);
if (ret < 0)
isert_err("Cmd: %p failed to post RDMA res\n", cmd);
diff --git a/drivers/infiniband/ulp/isert/ib_isert.h b/drivers/infiniband/ulp/isert/ib_isert.h
index d6fd248320ae..3b296bac4f60 100644
--- a/drivers/infiniband/ulp/isert/ib_isert.h
+++ b/drivers/infiniband/ulp/isert/ib_isert.h
@@ -126,6 +126,7 @@ struct isert_cmd {
struct rdma_rw_ctx rw;
struct work_struct comp_work;
struct scatterlist sg;
+ bool ctx_init_done;
};
static inline struct isert_cmd *tx_desc_to_cmd(struct iser_tx_desc *desc)
--
2.14.1
next prev parent reply other threads:[~2018-03-08 4:59 UTC|newest]
Thread overview: 71+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-03-08 4:57 [PATCH AUTOSEL for 4.14 01/67] Bluetooth: hci_qca: Avoid setup failure on missing rampatch Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 03/67] cpufreq: longhaul: Revert transition_delay_us to 200 ms Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 02/67] Bluetooth: btqcomsmd: Fix skb double free corruption Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 04/67] dt-bindings: net: add TI CC2560 Bluetooth chip Sasha Levin
2018-03-08 16:09 ` David Lechner
2018-03-09 16:38 ` Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 05/67] media: c8sectpfe: fix potential NULL pointer dereference in c8sectpfe_timer_interrupt Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 06/67] drm/msm: fix leak in failed get_pages Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 07/67] net: fec: add phy_reset_after_clk_enable() support Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 08/67] dm: ensure bio submission follows a depth-first tree walk Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 09/67] RDMA/iwpm: Fix uninitialized error code in iwpm_send_mapinfo() Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 10/67] hv_netvsc: Fix the receive buffer size limit Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 11/67] hv_netvsc: Fix the TX/RX buffer default sizes Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 12/67] KVM: x86: add support for emulating UMIP Sasha Levin
2018-03-08 6:46 ` Paolo Bonzini
2018-03-19 15:22 ` Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 14/67] rtlwifi: rtl_pci: Fix the bug when inactiveps is enabled Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 13/67] spi: sh-msiof: Avoid writing to registers from spi_master.setup() Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 17/67] ath10k: handling qos at STA side based on AP WMM enable/disable Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 15/67] rtlwifi: always initialize variables given to RT_TRACE() Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 16/67] media: bt8xx: Fix err 'bt878_probe()' Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 18/67] media: [RESEND] media: dvb-frontends: Add delay to Si2168 restart Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 19/67] qmi_wwan: set FLAG_SEND_ZLP to avoid network initiated disconnect Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 20/67] tty: goldfish: Enable 'earlycon' only if built-in Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 21/67] serial: 8250_dw: Disable clock on error Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 24/67] watchdog: Fix kref imbalance seen if handle_boot_enabled=0 Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 22/67] cros_ec: fix nul-termination for firmware build info Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 23/67] watchdog: Fix potential kref imbalance when opening watchdog Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 25/67] platform/chrome: Use proper protocol transfer function Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 26/67] dmaengine: zynqmp_dma: Fix race condition in the probe Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 27/67] drm/tilcdc: ensure nonatomic iowrite64 is not used Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 28/67] mmc: avoid removing non-removable hosts during suspend Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 30/67] /dev/mem: Add bounce buffer for copy-out Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 29/67] mmc: block: fix logical error to avoid memory leak Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 31/67] net: phy: meson-gxl: check phy_write return value Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 32/67] sfp: fix EEPROM reading in the case of non-SFF8472 SFPs Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 33/67] sfp: fix non-detection of PHY Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 34/67] media: s5p-mfc: Fix lock contention - request_firmware() once Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 35/67] rtc: ac100: Fix multiple race conditions Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 37/67] RDMA/cma: Use correct size when writing netlink stats Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 36/67] IB/ipoib: Avoid memory leak if the SA returns a different DGID Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 38/67] IB/umem: Fix use of npages/nmap fields Sasha Levin
2018-03-08 4:57 ` Sasha Levin [this message]
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 42/67] mmc: sdhci-xenon: wait 5ms after set 1.8V signal enable Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 40/67] vgacon: Set VGA struct resource types Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 41/67] omapdrm: panel: fix compatible vendor string for td028ttec1 Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 44/67] pty: cancel pty slave port buf's work in tty_release Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 43/67] drm/omap: DMM: Check for DMM readiness after successful transaction commit Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 46/67] PCI: designware-ep: Fix ->get_msi() to check MSI_EN bit Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 45/67] coresight: Fix disabling of CoreSight TPIU Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 48/67] PCI: rcar: Handle rcar_pcie_parse_request_of_pci_ranges() failures Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 49/67] media: davinci: fix a debug printk Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 47/67] PCI: endpoint: Fix find_first_zero_bit() usage Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 51/67] pinctrl: rockchip: enable clock when reading pin direction register Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 50/67] pinctrl: Really force states during suspend/resume Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 53/67] ip6_vti: adjust vti mtu according to mtu of lower device Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 52/67] iommu/vt-d: clean up pr_irq if request_threaded_irq fails Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 54/67] ip_gre: fix error path when erspan_rcv failed Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 55/67] ip_gre: fix potential memory leak in erspan_rcv Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 56/67] soc: qcom: smsm: fix child-node lookup Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 58/67] scsi: lpfc: Fix issues connecting with nvme initiator Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 57/67] scsi: lpfc: Fix SCSI LUN discovery when SCSI and NVME enabled Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 59/67] RDMA/ocrdma: Fix permissions for OCRDMA_RESET_STATS Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 60/67] ARM: dts: aspeed-evb: Add unit name to memory node Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 61/67] nfsd4: permit layoutget of executable-only files Sasha Levin
2018-03-08 4:57 ` [PATCH AUTOSEL for 4.14 62/67] clk: at91: pmc: Wait for clocks when resuming Sasha Levin
2018-03-08 4:58 ` [PATCH AUTOSEL for 4.14 63/67] clk: Don't touch hardware when reparenting during registration Sasha Levin
2018-03-08 4:58 ` [PATCH AUTOSEL for 4.14 64/67] clk: axi-clkgen: Correctly handle nocount bit in recalc_rate() Sasha Levin
2018-03-08 4:58 ` [PATCH AUTOSEL for 4.14 65/67] clk: si5351: Rename internal plls to avoid name collisions Sasha Levin
2018-03-08 4:58 ` [PATCH AUTOSEL for 4.14 67/67] dmaengine: ti-dma-crossbar: Fix event mapping for TPCC_EVT_MUX_60_63 Sasha Levin
2018-03-08 4:58 ` [PATCH AUTOSEL for 4.14 66/67] crypto: artpec6 - set correct iv size for gcm(aes) Sasha Levin
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=20180308045641.7814-39-alexander.levin@microsoft.com \
--to=alexander.levin@microsoft.com \
--cc=bharat@chelsio.com \
--cc=jgg@mellanox.com \
--cc=linux-kernel@vger.kernel.org \
--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