Revision c0cf4512a31eb3cec70b066bc36ed55f7d05b8c0 authored by Bart Van Assche on 23 June 2016, 07:35:48 UTC, committed by Doug Ledford on 23 June 2016, 16:04:09 UTC
The memory needed for the send and receive queues associated with
a QP is proportional to the max_sge parameter. The current value
of that parameter is such that with an mlx4 HCA the QP buffer size
is 8 MB. Since DMA is used for communication between HCA and CPU
that buffer either has to be allocated coherently or map_single()
must succeed for that buffer. Since large contiguous allocations
are fragile and since the maximum segment size for e.g. swiotlb
is 256 KB, reduce the max_sge parameter. This patch avoids that
the following text appears on the console after SRP logout and
relogin on a system equipped with multiple IB HCAs:

mlx4_core 0000:05:00.0: swiotlb buffer is full (sz: 8388608 bytes)
swiotlb: coherent allocation failed for device 0000:05:00.0 size=8388608
CPU: 11 PID: 148 Comm: kworker/11:1 Not tainted 4.7.0-rc4-dbg+ #1
Call Trace:
 [<ffffffff812c6d35>] dump_stack+0x67/0x92
 [<ffffffff812efe71>] swiotlb_alloc_coherent+0x141/0x150
 [<ffffffff810458be>] x86_swiotlb_alloc_coherent+0x3e/0x50
 [<ffffffffa03861fa>] mlx4_buf_direct_alloc.isra.5+0x9a/0x120 [mlx4_core]
 [<ffffffffa0386545>] mlx4_buf_alloc+0x165/0x1a0 [mlx4_core]
 [<ffffffffa035053d>] create_qp_common.isra.29+0x57d/0xff0 [mlx4_ib]
 [<ffffffffa03510da>] mlx4_ib_create_qp+0x12a/0x3f0 [mlx4_ib]
 [<ffffffffa031154a>] ib_create_qp+0x3a/0x250 [ib_core]
 [<ffffffffa055dd4b>] srpt_cm_handler+0x4bb/0xcad [ib_srpt]
 [<ffffffffa02c1ab0>] cm_process_work+0x20/0xf0 [ib_cm]
 [<ffffffffa02c3640>] cm_work_handler+0x1ac0/0x2059 [ib_cm]
 [<ffffffff810737ed>] process_one_work+0x19d/0x490
 [<ffffffff81073b29>] worker_thread+0x49/0x490
 [<ffffffff8107a0ea>] kthread+0xea/0x100
 [<ffffffff815b25af>] ret_from_fork+0x1f/0x40

Fixes: b99f8e4d7bcd ("IB/srpt: convert to the generic RDMA READ/WRITE API")
Signed-off-by: Bart Van Assche <bart.vanassche@sandisk.com>
Cc: Laurence Oberman <loberman@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
1 parent 37e07cd
Raw File
netdev-notifier-error-inject.c
#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/netdevice.h>

#include "notifier-error-inject.h"

static int priority;
module_param(priority, int, 0);
MODULE_PARM_DESC(priority, "specify netdevice notifier priority");

static struct notifier_err_inject netdev_notifier_err_inject = {
	.actions = {
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_REGISTER) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_CHANGEMTU) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_CHANGENAME) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_PRE_UP) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_PRE_TYPE_CHANGE) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_POST_INIT) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_PRECHANGEMTU) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_PRECHANGEUPPER) },
		{ NOTIFIER_ERR_INJECT_ACTION(NETDEV_CHANGEUPPER) },
		{}
	}
};

static struct dentry *dir;

static int netdev_err_inject_init(void)
{
	int err;

	dir = notifier_err_inject_init("netdev", notifier_err_inject_dir,
				       &netdev_notifier_err_inject, priority);
	if (IS_ERR(dir))
		return PTR_ERR(dir);

	err = register_netdevice_notifier(&netdev_notifier_err_inject.nb);
	if (err)
		debugfs_remove_recursive(dir);

	return err;
}

static void netdev_err_inject_exit(void)
{
	unregister_netdevice_notifier(&netdev_notifier_err_inject.nb);
	debugfs_remove_recursive(dir);
}

module_init(netdev_err_inject_init);
module_exit(netdev_err_inject_exit);

MODULE_DESCRIPTION("Netdevice notifier error injection module");
MODULE_LICENSE("GPL");
MODULE_AUTHOR("Nikolay Aleksandrov <razor@blackwall.org>");
back to top