patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Maxime Coquelin <maxime.coquelin@redhat.com>
To: "Xia, Chenbo" <chenbo.xia@intel.com>,
	"dev@dpdk.org" <dev@dpdk.org>,
	"amorenoz@redhat.com" <amorenoz@redhat.com>
Cc: "stable@dpdk.org" <stable@dpdk.org>
Subject: Re: [dpdk-stable] [PATCH 1/7] vhost: fix virtqueues metadata allocation
Date: Wed, 21 Oct 2020 14:06:46 +0200
Message-ID: <cdfd2baf-88dc-a7b8-1df9-96bca4404871@redhat.com> (raw)
In-Reply-To: <MN2PR11MB40632B50C26BB548CD94F8689C1C0@MN2PR11MB4063.namprd11.prod.outlook.com>

Hi Chenbon

On 10/21/20 1:10 PM, Xia, Chenbo wrote:
> Hi Maxime,
> 
>> -----Original Message-----
>> From: Maxime Coquelin <maxime.coquelin@redhat.com>
>> Sent: Tuesday, October 20, 2020 1:34 AM
>> To: dev@dpdk.org; Xia, Chenbo <chenbo.xia@intel.com>; amorenoz@redhat.com
>> Cc: Maxime Coquelin <maxime.coquelin@redhat.com>; stable@dpdk.org
>> Subject: [PATCH 1/7] vhost: fix virtqueues metadata allocation
>>
>> The Vhost-user backend implementation assumes there will be
>> no holes in the device's array of virtqueues metadata
>> pointers.
>>
>> It can happen though, and would cause segmentation faults,
>> memory leaks or undefined behaviour.
> 
> Could I ask when will this happen? 
> 
> When QEMU does not configure all virtqueues? I'm not very sure.
> Could you point that out for me?

It has been reported by our QE when doing reconnect with multiqueue with
vIOMMU enabled:
https://bugzilla.redhat.com/show_bug.cgi?id=1880299

Regards,
Maxime

> Thanks!
> Chenbo
> 
>>
>> This patch keep the assumption that there is no holes in this
>> array, and allocate all uninitialized virtqueues metadata up
>> to requested index.
>>
>> Fixes: 160cbc815b41 ("vhost: remove a hack on queue allocation")
>> Cc: stable@dpdk.org
>>
>> Suggested-by: Adrian Moreno <amorenoz@redhat.com>
>> Signed-off-by: Maxime Coquelin <maxime.coquelin@redhat.com>
>> ---
>>  lib/librte_vhost/vhost.c | 33 ++++++++++++++++++++-------------
>>  1 file changed, 20 insertions(+), 13 deletions(-)
>>
>> diff --git a/lib/librte_vhost/vhost.c b/lib/librte_vhost/vhost.c
>> index 6068c38ec6..0c9ba3b3af 100644
>> --- a/lib/librte_vhost/vhost.c
>> +++ b/lib/librte_vhost/vhost.c
>> @@ -579,22 +579,29 @@ int
>>  alloc_vring_queue(struct virtio_net *dev, uint32_t vring_idx)
>>  {
>>  	struct vhost_virtqueue *vq;
>> +	uint32_t i;
>>
>> -	vq = rte_malloc(NULL, sizeof(struct vhost_virtqueue), 0);
>> -	if (vq == NULL) {
>> -		VHOST_LOG_CONFIG(ERR,
>> -			"Failed to allocate memory for vring:%u.\n", vring_idx);
>> -		return -1;
>> -	}
>> +	/* Also allocate holes, if any, up to requested vring index. */
>> +	for (i = 0; i <= vring_idx; i++) {
>> +		if (dev->virtqueue[i])
>> +			continue;
>>
>> -	dev->virtqueue[vring_idx] = vq;
>> -	init_vring_queue(dev, vring_idx);
>> -	rte_spinlock_init(&vq->access_lock);
>> -	vq->avail_wrap_counter = 1;
>> -	vq->used_wrap_counter = 1;
>> -	vq->signalled_used_valid = false;
>> +		vq = rte_malloc(NULL, sizeof(struct vhost_virtqueue), 0);
>> +		if (vq == NULL) {
>> +			VHOST_LOG_CONFIG(ERR,
>> +				"Failed to allocate memory for vring:%u.\n", i);
>> +			return -1;
>> +		}
>> +
>> +		dev->virtqueue[i] = vq;
>> +		init_vring_queue(dev, vring_idx);
>> +		rte_spinlock_init(&vq->access_lock);
>> +		vq->avail_wrap_counter = 1;
>> +		vq->used_wrap_counter = 1;
>> +		vq->signalled_used_valid = false;
>> +	}
>>
>> -	dev->nr_vring += 1;
>> +	dev->nr_vring = RTE_MAX(dev->nr_vring, vring_idx + 1);
>>
>>  	return 0;
>>  }
>> --
>> 2.26.2
> 


  reply	other threads:[~2020-10-21 12:07 UTC|newest]

Thread overview: 14+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <20201019173415.582407-1-maxime.coquelin@redhat.com>
2020-10-19 17:34 ` Maxime Coquelin
2020-10-21 11:10   ` Xia, Chenbo
2020-10-21 12:06     ` Maxime Coquelin [this message]
2020-10-22 11:00       ` Xia, Chenbo
2020-10-19 17:34 ` [dpdk-stable] [PATCH 2/7] vhost: validate index in available entries API Maxime Coquelin
2020-10-21 11:28   ` Xia, Chenbo
2020-10-19 17:34 ` [dpdk-stable] [PATCH 3/7] vhost: validate index in guest notification API Maxime Coquelin
2020-10-21 11:30   ` Xia, Chenbo
2020-10-19 17:34 ` [dpdk-stable] [PATCH 4/7] vhost: validate index in live-migration API Maxime Coquelin
2020-10-21 11:30   ` Xia, Chenbo
2020-10-19 17:34 ` [dpdk-stable] [PATCH 5/7] vhost: validate index in inflight API Maxime Coquelin
2020-10-21 11:30   ` Xia, Chenbo
2020-10-19 17:34 ` [dpdk-stable] [PATCH 6/7] vhost: validate index in async API Maxime Coquelin
2020-10-21 11:31   ` Xia, Chenbo

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=cdfd2baf-88dc-a7b8-1df9-96bca4404871@redhat.com \
    --to=maxime.coquelin@redhat.com \
    --cc=amorenoz@redhat.com \
    --cc=chenbo.xia@intel.com \
    --cc=dev@dpdk.org \
    --cc=stable@dpdk.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

patches for DPDK stable branches

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/stable/0 stable/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 stable stable/ https://inbox.dpdk.org/stable \
		stable@dpdk.org
	public-inbox-index stable

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.stable


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git