From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
stable@vger.kernel.org, Carol L Soto <clsoto@linux.vnet.ibm.com>,
Jack Morgenstein <jackm@dev.mellanox.co.il>,
Or Gerlitz <ogerlitz@mellanox.com>,
"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 3.14 05/37] net/mlx4: Copy/set only sizeof struct mlx4_eqe bytes
Date: Mon, 7 Dec 2015 09:26:17 -0500 [thread overview]
Message-ID: <20151207141743.511219221@linuxfoundation.org> (raw)
In-Reply-To: <20151207141743.221453847@linuxfoundation.org>
3.14-stable review patch. If anyone has any objections, please let me know.
------------------
From: Carol L Soto <clsoto@linux.vnet.ibm.com>
[ Upstream commit c02b05011fadf8e409e41910217ca689f2fc9d91 ]
When doing memcpy/memset of EQEs, we should use sizeof struct
mlx4_eqe as the base size and not caps.eqe_size which could be bigger.
If caps.eqe_size is bigger than the struct mlx4_eqe then we corrupt
data in the master context.
When using a 64 byte stride, the memcpy copied over 63 bytes to the
slave_eq structure. This resulted in copying over the entire eqe of
interest, including its ownership bit -- and also 31 bytes of garbage
into the next WQE in the slave EQ -- which did NOT include the ownership
bit (and therefore had no impact).
However, once the stride is increased to 128, we are overwriting the
ownership bits of *three* eqes in the slave_eq struct. This results
in an incorrect ownership bit for those eqes, which causes the eq to
seem to be full. The issue therefore surfaced only once 128-byte EQEs
started being used in SRIOV and (overarchitectures that have 128/256
byte cache-lines such as PPC) - e.g after commit 77507aa249ae
"net/mlx4_core: Enable CQE/EQE stride support".
Fixes: 08ff32352d6f ('mlx4: 64-byte CQE/EQE support')
Signed-off-by: Carol L Soto <clsoto@linux.vnet.ibm.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
drivers/net/ethernet/mellanox/mlx4/cmd.c | 2 +-
drivers/net/ethernet/mellanox/mlx4/eq.c | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
--- a/drivers/net/ethernet/mellanox/mlx4/cmd.c
+++ b/drivers/net/ethernet/mellanox/mlx4/cmd.c
@@ -1987,7 +1987,7 @@ int mlx4_multi_func_init(struct mlx4_dev
spin_lock_init(&s_state->lock);
}
- memset(&priv->mfunc.master.cmd_eqe, 0, dev->caps.eqe_size);
+ memset(&priv->mfunc.master.cmd_eqe, 0, sizeof(struct mlx4_eqe));
priv->mfunc.master.cmd_eqe.type = MLX4_EVENT_TYPE_CMD;
INIT_WORK(&priv->mfunc.master.comm_work,
mlx4_master_comm_channel);
--- a/drivers/net/ethernet/mellanox/mlx4/eq.c
+++ b/drivers/net/ethernet/mellanox/mlx4/eq.c
@@ -183,7 +183,7 @@ static void slave_event(struct mlx4_dev
return;
}
- memcpy(s_eqe, eqe, dev->caps.eqe_size - 1);
+ memcpy(s_eqe, eqe, sizeof(struct mlx4_eqe) - 1);
s_eqe->slave_id = slave;
/* ensure all information is written before setting the ownersip bit */
wmb();
next prev parent reply other threads:[~2015-12-07 14:34 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-12-07 14:26 [PATCH 3.14 00/37] 3.14.58-stable review Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 02/37] macvtap: unbreak receiving of gro skb with frag list Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 03/37] ppp: fix pppoe_dev deletion condition in pppoe_release() Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 04/37] RDS-TCP: Recover correctly from pskb_pull()/pksb_trim() failure in rds_tcp_data_recv Greg Kroah-Hartman
2015-12-07 14:26 ` Greg Kroah-Hartman [this message]
2015-12-07 14:26 ` [PATCH 3.14 06/37] stmmac: Correctly report PTP capabilities Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 07/37] ipmr: fix possible race resulting from improper usage of IP_INC_STATS_BH() in preemptible context Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 08/37] net: avoid NULL deref in inet_ctl_sock_destroy() Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 09/37] packet: race condition in packet_bind Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 10/37] net: fix a race in dst_release() Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 11/37] virtio-net: drop NETIF_F_FRAGLIST Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 12/37] RDS: verify the underlying transport exists before creating a connection Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 13/37] ARM: 8426/1: dma-mapping: add missing range check in dma_mmap() Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 14/37] ARM: 8427/1: dma-mapping: add support for offset parameter " Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 15/37] ARM: orion: Fix DSA platform device after mvmdio conversion Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 16/37] ARM: common: edma: Fix channel parameter for irq callbacks Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 17/37] x86/setup: Extend low identity map to cover whole kernel range Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 18/37] x86/setup: Fix low identity map for >= 2GB " Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 19/37] x86/cpu: Call verify_cpu() after having entered long mode too Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 20/37] x86/cpu: Fix SMAP check in PVOPS environments Greg Kroah-Hartman
2015-12-07 14:26 ` Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 21/37] mac80211: fix driver RSSI event calculations Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 22/37] net: mvneta: Fix CPU_MAP registers initialisation Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 23/37] mwifiex: fix mwifiex_rdeeprom_read() Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 24/37] staging: rtl8712: Add device ID for Sitecom WLA2100 Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 25/37] Bluetooth: hidp: fix device disconnect on idle timeout Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 26/37] Bluetooth: ath3k: Add new AR3012 0930:021c id Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 27/37] Bluetooth: ath3k: Add support of AR3012 0cf3:817b device Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 28/37] can: sja1000: clear interrupts on start Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 29/37] arm64: Fix compat register mappings Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 30/37] usblp: do not set TASK_INTERRUPTIBLE before lock Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 32/37] USB: ti_usb_3410_5052: Add Honeywell HGI80 ID Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 35/37] ALSA: usb-audio: add packet size quirk for the Medeli DD305 Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 36/37] ALSA: usb-audio: prevent CH345 multiport output SysEx corruption Greg Kroah-Hartman
2015-12-07 14:26 ` [PATCH 3.14 37/37] ALSA: usb-audio: work around CH345 input " Greg Kroah-Hartman
2015-12-07 17:18 ` [PATCH 3.14 00/37] 3.14.58-stable review Shuah Khan
2015-12-07 21:16 ` Guenter Roeck
2015-12-09 3:26 ` 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=20151207141743.511219221@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=clsoto@linux.vnet.ibm.com \
--cc=davem@davemloft.net \
--cc=jackm@dev.mellanox.co.il \
--cc=linux-kernel@vger.kernel.org \
--cc=ogerlitz@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 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.