All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jason Gunthorpe <jgg@nvidia.com>
To: Leon Romanovsky <leon@kernel.org>
Cc: Doug Ledford <dledford@redhat.com>,
	Mark Zhang <markz@mellanox.com>,
	Haggai Eran <haggaie@mellanox.com>, <linux-rdma@vger.kernel.org>,
	"Maor Gottlieb" <maorg@mellanox.com>
Subject: Re: [PATCH rdma-rc] RDMA/cma: Protect bind_list and listen_list while finding matching cm id
Date: Thu, 18 Jun 2020 09:56:13 -0300	[thread overview]
Message-ID: <20200618125613.GA2405869@mellanox.com> (raw)
In-Reply-To: <20200616104304.2426081-1-leon@kernel.org>

On Tue, Jun 16, 2020 at 01:43:04PM +0300, Leon Romanovsky wrote:
> From: Mark Zhang <markz@mellanox.com>
> 
> Add missing lock when the bind_list and listen_list must be accessed
> with lock. In addition add lockdep asserts to verify that lock is held.
> 
> [14127.352448] general protection fault: 0000 [#1] SMP NOPTI
> [14127.356017] CPU: 226 PID: 126135 Comm: kworker/226:1 Tainted: G OE 4.12.14-150.47-default #1 SLE15
> [14127.363169] Hardware name: Cray Inc. Windom/Windom, BIOS 0.8.7 01-10-2020
> [14127.368545] Workqueue: ib_cm cm_work_handler [ib_cm]
> [14127.373185] task: ffff9c5a60a1d2c0 task.stack: ffffc1d91f554000
> [14127.379151] RIP: 0010:cma_ib_req_handler+0x3f1/0x11b0 [rdma_cm]
> [14127.382316] RSP: 0018:ffffc1d91f557b40 EFLAGS: 00010286
> [14127.385693] RAX: deacffffffffff30 RBX: 0000000000000001 RCX: ffff9c2af5bb6000
> [14127.391431] RDX: 00000000000000a9 RSI: ffff9c5aa4ed2f10 RDI: ffffc1d91f557b08
> [14127.396263] RBP: ffffc1d91f557d90 R08: ffff9c340cc80000 R09: ffff9c2c0f901900
> [14127.401531] R10: 0000000000000000 R11: 0000000000000001 R12: deacffffffffff30
> [14127.410279] R13: ffff9c5a48aeec00 R14: ffffc1d91f557c30 R15: ffff9c5c2eea3688
> [14127.418274] FS: 0000000000000000(0000) GS:ffff9c5c2fa80000(0000) knlGS:0000000000000000
> [14127.427259] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [14127.432108] CR2: 00002b5cc03fa320 CR3: 0000003f8500a000 CR4: 00000000003406e0
> [14127.438461] Call Trace:
> [14127.441557] ? rdma_addr_cancel+0xa0/0xa0 [ib_core]
> [14127.447922] ? cm_process_work+0x28/0x140 [ib_cm]
> [14127.454877] cm_process_work+0x28/0x140 [ib_cm]
> [14127.458933] ? cm_get_bth_pkey.isra.44+0x34/0xa0 [ib_cm]
> [14127.464246] cm_work_handler+0xa06/0x1a6f [ib_cm]
> [14127.470069] ? __switch_to_asm+0x34/0x70
> [14127.477089] ? __switch_to_asm+0x34/0x70
> [14127.484283] ? __switch_to_asm+0x40/0x70
> [14127.491010] ? __switch_to_asm+0x34/0x70
> [14127.495557] ? __switch_to_asm+0x40/0x70
> [14127.498530] ? __switch_to_asm+0x34/0x70
> [14127.502205] ? __switch_to_asm+0x40/0x70
> [14127.508609] ? __switch_to+0x7c/0x4b0
> [14127.512301] ? __switch_to_asm+0x40/0x70
> [14127.516245] ? __switch_to_asm+0x34/0x70
> [14127.518631] process_one_work+0x1da/0x400
> [14127.522672] worker_thread+0x2b/0x3f0
> [14127.527426] ? process_one_work+0x400/0x400
> [14127.534769] kthread+0x118/0x140
> [14127.538181] ? kthread_create_on_node+0x40/0x40
> [14127.544509] ret_from_fork+0x22/0x40
> [14127.548842] Code: 00 66 83 f8 02 0f 84 ca 05 00 00 49 8b 84 24 d0 01 00 00 48 85 c0 0f 84 68 07 00 00 48 2d d0 01
> 00 00 49 89 c4 0f 84 59 07 00 00 <41> 0f b7 44 24 20 49 8b 77 50 66 83 f8 0a 75 9e 49 8b 7c 24 28
> [14127.573512] Modules linked in: squashfs lz4_decompress loop rpcsec_gss_krb5 auth_rpcgss nfsv4 dns_resolver nfs lockd grace sunrpc fscache 8021q garp mrp stp llc af_packet iscsi_ibft iscsi_boot_sysfs rdma_ucm(OEX) ib_ucm(OEX) rdma_cm(OEX) iw_cm(OEX) configfs ib_ipoib(OEX) ib_cm(OEX) ib_umad(OEX) mlx5_fpga_tools(OEX) mlx4_en(OEX) mlx4_ib(OEX) mlx4_core(OEX) mlx5_ib(OEX) ib_uverbs(OEX) edac_mce_amd kvm_amd ib_core(OEX) kvm irqbypass crc32_pclmul crc32c_intel ghash_clmulni_intel pcbc aesni_intel aes_x86_64 crypto_simd glue_helper cryptd pcspkr mlx5_core(OEX) mlxfw(OEX) devlink vfio_mdev(OEX) igb vfio_iommu_type1 ptp vfio ahci mdev(OEX) libahci xhci_pci pps_core xhci_hcd i2c_algo_bit libata mlx_compat(OEX) ccp dca usbcore i2c_piix4 shpchp pcc_cpufreq acpi_cpufreq button sg scsi_mod efivarfs autofs4
> 
> Fixes: 4c21b5bcef73 ("IB/cma: Add net_dev and private data checks to RDMA CM")
> Signed-off-by: Mark Zhang <markz@mellanox.com>
> Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
> Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
> ---
>  drivers/infiniband/core/cma.c | 18 ++++++++++++++++++
>  1 file changed, 18 insertions(+)

Applied to for-rc, thanks

Jason

      reply	other threads:[~2020-06-18 12:59 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-06-16 10:43 [PATCH rdma-rc] RDMA/cma: Protect bind_list and listen_list while finding matching cm id Leon Romanovsky
2020-06-18 12:56 ` Jason Gunthorpe [this message]

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=20200618125613.GA2405869@mellanox.com \
    --to=jgg@nvidia.com \
    --cc=dledford@redhat.com \
    --cc=haggaie@mellanox.com \
    --cc=leon@kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=maorg@mellanox.com \
    --cc=markz@mellanox.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.