linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org,
	Jack Morgenstein <jackm@dev.mellanox.co.il>,
	Tariq Toukan <tariqt@mellanox.com>,
	"David S. Miller" <davem@davemloft.net>,
	Sasha Levin <alexander.levin@verizon.com>
Subject: [PATCH 4.4 28/46] net/mlx4_core: Avoid command timeouts during VF driver device shutdown
Date: Thu, 15 Jun 2017 19:52:47 +0200	[thread overview]
Message-ID: <20170615175219.608632228@linuxfoundation.org> (raw)
In-Reply-To: <20170615175218.286057711@linuxfoundation.org>

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Jack Morgenstein <jackm@dev.mellanox.co.il>


[ Upstream commit d585df1c5ccf995fcee910705ad7a9cdd11d4152 ]

Some Hypervisors detach VFs from VMs by instantly causing an FLR event
to be generated for a VF.

In the mlx4 case, this will cause that VF's comm channel to be disabled
before the VM has an opportunity to invoke the VF device's "shutdown"
method.

The result is that the VF driver on the VM will experience a command
timeout during the shutdown process when the Hypervisor does not deliver
a command-completion event to the VM.

To avoid FW command timeouts on the VM when the driver's shutdown method
is invoked, we detect the absence of the VF's comm channel at the very
start of the shutdown process. If the comm-channel has already been
disabled, we cause all FW commands during the device shutdown process to
immediately return success (and thus avoid all command timeouts).

Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/net/ethernet/mellanox/mlx4/catas.c |    2 +-
 drivers/net/ethernet/mellanox/mlx4/intf.c  |   12 ++++++++++++
 drivers/net/ethernet/mellanox/mlx4/mlx4.h  |    1 +
 3 files changed, 14 insertions(+), 1 deletion(-)

--- a/drivers/net/ethernet/mellanox/mlx4/catas.c
+++ b/drivers/net/ethernet/mellanox/mlx4/catas.c
@@ -158,7 +158,7 @@ static int mlx4_reset_slave(struct mlx4_
 	return -ETIMEDOUT;
 }
 
-static int mlx4_comm_internal_err(u32 slave_read)
+int mlx4_comm_internal_err(u32 slave_read)
 {
 	return (u32)COMM_CHAN_EVENT_INTERNAL_ERR ==
 		(slave_read & (u32)COMM_CHAN_EVENT_INTERNAL_ERR) ? 1 : 0;
--- a/drivers/net/ethernet/mellanox/mlx4/intf.c
+++ b/drivers/net/ethernet/mellanox/mlx4/intf.c
@@ -218,6 +218,18 @@ void mlx4_unregister_device(struct mlx4_
 	struct mlx4_interface *intf;
 
 	mlx4_stop_catas_poll(dev);
+	if (dev->persist->interface_state & MLX4_INTERFACE_STATE_DELETION &&
+	    mlx4_is_slave(dev)) {
+		/* In mlx4_remove_one on a VF */
+		u32 slave_read =
+			swab32(readl(&mlx4_priv(dev)->mfunc.comm->slave_read));
+
+		if (mlx4_comm_internal_err(slave_read)) {
+			mlx4_dbg(dev, "%s: comm channel is down, entering error state.\n",
+				 __func__);
+			mlx4_enter_error_state(dev->persist);
+		}
+	}
 	mutex_lock(&intf_mutex);
 
 	list_for_each_entry(intf, &intf_list, list)
--- a/drivers/net/ethernet/mellanox/mlx4/mlx4.h
+++ b/drivers/net/ethernet/mellanox/mlx4/mlx4.h
@@ -1205,6 +1205,7 @@ void mlx4_qp_event(struct mlx4_dev *dev,
 void mlx4_srq_event(struct mlx4_dev *dev, u32 srqn, int event_type);
 
 void mlx4_enter_error_state(struct mlx4_dev_persistent *persist);
+int mlx4_comm_internal_err(u32 slave_read);
 
 int mlx4_SENSE_PORT(struct mlx4_dev *dev, int port,
 		    enum mlx4_port_type *type);

  parent reply	other threads:[~2017-06-15 17:55 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-06-15 17:52 [PATCH 4.4 00/46] 4.4.73-stable review Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 01/46] s390/vmem: fix identity mapping Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 02/46] partitions/msdos: FreeBSD UFS2 file systems are not recognized Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 03/46] ARM: dts: imx6dl: Fix the VDD_ARM_CAP voltage for 396MHz operation Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 04/46] staging: rtl8192e: rtl92e_fill_tx_desc fix write to mapped out memory Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 05/46] Call echo service immediately after socket reconnect Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 06/46] net: xilinx_emaclite: fix freezes due to unordered I/O Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 07/46] net: xilinx_emaclite: fix receive buffer overflow Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 08/46] ipv6: Handle IPv4-mapped src to in6addr_any dst Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 09/46] ipv6: Inhibit IPv4-mapped src address on the wire Greg Kroah-Hartman
2017-06-29 12:13   ` Ben Hutchings
2017-06-29 12:35     ` Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 10/46] NET: Fix /proc/net/arp for AX.25 Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 11/46] NET: mkiss: Fix panic Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 12/46] net: hns: Fix the device being used for dma mapping during TX Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 15/46] i2c: piix4: Fix request_region size Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 17/46] PM / runtime: Avoid false-positive warnings from might_sleep_if() Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 18/46] jump label: pass kbuild_cflags when checking for asm goto support Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 19/46] kasan: respect /proc/sys/kernel/traceoff_on_warning Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 20/46] log2: make order_base_2() behave correctly on const input value zero Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 21/46] ethtool: do not vzalloc(0) on registers dump Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 22/46] fscache: Fix dead object requeue Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 23/46] fscache: Clear outstanding writes when disabling a cookie Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 24/46] FS-Cache: Initialise stores_lock in netfs cookie Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 25/46] ipv6: fix flow labels when the traffic class is non-0 Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 26/46] drm/nouveau: prevent userspace from deleting client object Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 27/46] drm/nouveau/fence/g84-: protect against concurrent access to semaphore buffers Greg Kroah-Hartman
2017-06-15 17:52 ` Greg Kroah-Hartman [this message]
2017-06-15 17:52 ` [PATCH 4.4 29/46] gianfar: synchronize DMA API usage by free_skb_rx_queue w/ gfar_new_page Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 30/46] pinctrl: berlin-bg4ct: fix the value for "sd1a" of pin SCRD0_CRD_PRES Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 31/46] net: adaptec: starfire: add checks for dma mapping errors Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 32/46] parisc, parport_gsc: Fixes for printk continuation lines Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 33/46] drm/nouveau: Dont enabling polling twice on runtime resume Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 34/46] drm/ast: Fixed system hanged if disable P2A Greg Kroah-Hartman
2017-06-29 13:45   ` Ben Hutchings
2017-07-03  7:27     ` Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 35/46] ravb: unmap descriptors when freeing rings Greg Kroah-Hartman
2017-06-29 13:58   ` Ben Hutchings
2017-07-03 12:52     ` Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 36/46] nfs: Fix "Dont increment lock sequence ID after NFS4ERR_MOVED" Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 37/46] r8152: re-schedule napi for tx Greg Kroah-Hartman
2017-06-29 14:23   ` Ben Hutchings
2017-06-15 17:52 ` [PATCH 4.4 38/46] r8152: fix rtl8152_post_reset function Greg Kroah-Hartman
2017-06-15 17:52 ` [PATCH 4.4 39/46] r8152: avoid start_xmit to schedule napi when napi is disabled Greg Kroah-Hartman
2017-06-29 14:35   ` Ben Hutchings
2017-06-15 17:52 ` [PATCH 4.4 40/46] sctp: sctp_addr_id2transport should verify the addr before looking up assoc Greg Kroah-Hartman
2017-06-15 17:53 ` [PATCH 4.4 41/46] romfs: use different way to generate fsid for BLOCK or MTD Greg Kroah-Hartman
2017-06-15 17:53 ` [PATCH 4.4 42/46] proc: add a schedule point in proc_pid_readdir() Greg Kroah-Hartman
2017-06-15 17:53 ` [PATCH 4.4 43/46] tipc: ignore requests when the connection state is not CONNECTED Greg Kroah-Hartman
2017-06-15 17:53 ` [PATCH 4.4 44/46] xtensa: dont use linux IRQ #0 Greg Kroah-Hartman
2017-06-15 17:53 ` [PATCH 4.4 45/46] s390/kvm: do not rely on the ILC on kvm host protection fauls Greg Kroah-Hartman
2017-06-15 17:53 ` [PATCH 4.4 46/46] sparc64: make string buffers large enough Greg Kroah-Hartman
2017-06-15 22:24 ` [PATCH 4.4 00/46] 4.4.73-stable review Shuah Khan
2017-06-16  0:39 ` Guenter Roeck

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=20170615175219.608632228@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=alexander.levin@verizon.com \
    --cc=davem@davemloft.net \
    --cc=jackm@dev.mellanox.co.il \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tariqt@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 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).