From: Paolo Bonzini <pbonzini@redhat.com>
To: "Richard W.M. Jones" <rjones@redhat.com>
Cc: Christoph Hellwig <hch@lst.de>,
linux-scsi@vger.kernel.org, linux-kernel@vger.kernel.org,
"Martin K. Petersen" <martin.petersen@oracle.com>
Subject: Re: Increased memory usage with scsi-mq
Date: Thu, 10 Aug 2017 16:30:54 +0200 [thread overview]
Message-ID: <367767f5-b3b2-6905-759d-d97196e7f140@redhat.com> (raw)
In-Reply-To: <20170810141634.GA20914@redhat.com>
On 10/08/2017 16:16, Richard W.M. Jones wrote:
> On Thu, Aug 10, 2017 at 02:53:58PM +0200, Paolo Bonzini wrote:
>> can_queue and cmd_per_lun are different. can_queue should be set to the
>> value of vq->vring.num where vq is the command virtqueue (the first one
>> is okay if there's >1).
>>
>> If you want to change it, you'll have to do so in QEMU.
>
> Here are a couple more patches I came up with, the first to Linux,
> the second to qemu.
>
> There are a few problems ...
>
> (1) Setting virtqueue_size to < 128 causes a BUG_ON failure in
> virtio_ring.c:virtqueue_add at:
>
> BUG_ON(total_sg > vq->vring.num);
That bug is bogus. You can change it to
WARN_ON_ONCE(total_sg > vq->vring.num && !vq->indirect);
and move it in the "else" here:
if (vq->indirect && total_sg > 1 && vq->vq.num_free)
desc = alloc_indirect(_vq, total_sg, gfp);
else
desc = NULL;
You just get a -ENOSPC after the WARN, so no need to crash the kernel!
> (2) Can/should the ctrl, event and cmd queue sizes be set to the same
> values? Or should ctrl & event be left at 128?
It's okay if they're all the same.
> (3) It seems like it might be a problem that virtqueue_size is not
> passed through the virtio_scsi_conf struct (which is ABI between the
> kernel and qemu AFAICT and so not easily modifiable). However I think
> this is not a problem because virtqueue size is stored in the
> Virtqueue Descriptor table, which is how the kernel gets the value.
> Is that right?
Yes.
Paolo
> Rich.
>
>
> --- kernel patch ---
>
> diff --git a/drivers/scsi/virtio_scsi.c b/drivers/scsi/virtio_scsi.c
> index 9be211d68b15..d6b4ff634c0d 100644
> --- a/drivers/scsi/virtio_scsi.c
> +++ b/drivers/scsi/virtio_scsi.c
> @@ -972,6 +972,8 @@ static int virtscsi_probe(struct virtio_device *vdev)
> if (err)
> goto virtscsi_init_failed;
>
> + shost->can_queue = virtqueue_get_vring_size(vscsi->req_vqs[0].vq);
> +
> cmd_per_lun = virtscsi_config_get(vdev, cmd_per_lun) ?: 1;
> shost->cmd_per_lun = min_t(u32, cmd_per_lun, shost->can_queue);
> shost->max_sectors = virtscsi_config_get(vdev, max_sectors) ?: 0xFFFF;
>
>
> --- qemu patch ---
>
> diff --git a/hw/scsi/virtio-scsi.c b/hw/scsi/virtio-scsi.c
> index eb639442d1..aadd99aad1 100644
> --- a/hw/scsi/virtio-scsi.c
> +++ b/hw/scsi/virtio-scsi.c
> @@ -867,10 +867,10 @@ void virtio_scsi_common_realize(DeviceState *dev,
> s->sense_size = VIRTIO_SCSI_SENSE_DEFAULT_SIZE;
> s->cdb_size = VIRTIO_SCSI_CDB_DEFAULT_SIZE;
>
> - s->ctrl_vq = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE, ctrl);
> - s->event_vq = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE, evt);
> + s->ctrl_vq = virtio_add_queue(vdev, s->conf.virtqueue_size, ctrl);
> + s->event_vq = virtio_add_queue(vdev, s->conf.virtqueue_size, evt);
> for (i = 0; i < s->conf.num_queues; i++) {
> - s->cmd_vqs[i] = virtio_add_queue(vdev, VIRTIO_SCSI_VQ_SIZE, cmd);
> + s->cmd_vqs[i] = virtio_add_queue(vdev, s->conf.virtqueue_size, cmd);
> }
> }
>
> @@ -917,6 +917,7 @@ static void virtio_scsi_device_unrealize(DeviceState *dev, Error **errp)
>
> static Property virtio_scsi_properties[] = {
> DEFINE_PROP_UINT32("num_queues", VirtIOSCSI, parent_obj.conf.num_queues, 1),
> + DEFINE_PROP_UINT32("virtqueue_size", VirtIOSCSI, parent_obj.conf.virtqueue_size, 128),
> DEFINE_PROP_UINT32("max_sectors", VirtIOSCSI, parent_obj.conf.max_sectors,
> 0xFFFF),
> DEFINE_PROP_UINT32("cmd_per_lun", VirtIOSCSI, parent_obj.conf.cmd_per_lun,
> diff --git a/include/hw/virtio/virtio-scsi.h b/include/hw/virtio/virtio-scsi.h
> index de6ae5a9f6..e30a92d3e7 100644
> --- a/include/hw/virtio/virtio-scsi.h
> +++ b/include/hw/virtio/virtio-scsi.h
> @@ -48,6 +48,7 @@ typedef struct virtio_scsi_config VirtIOSCSIConfig;
>
> struct VirtIOSCSIConf {
> uint32_t num_queues;
> + uint32_t virtqueue_size;
> uint32_t max_sectors;
> uint32_t cmd_per_lun;
> #ifdef CONFIG_VHOST_SCSI
>
>
next prev parent reply other threads:[~2017-08-10 14:30 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-04 21:00 Richard W.M. Jones
2017-08-05 8:44 ` Christoph Hellwig
2017-08-05 9:27 ` Richard W.M. Jones
2017-08-05 13:39 ` Christoph Hellwig
2017-08-05 15:51 ` Richard W.M. Jones
2017-08-07 12:11 ` Paolo Bonzini
2017-08-07 12:27 ` Richard W.M. Jones
2017-08-07 13:07 ` Paolo Bonzini
2017-08-09 16:01 ` Christoph Hellwig
2017-08-09 16:50 ` Paolo Bonzini
2017-08-10 12:22 ` Richard W.M. Jones
2017-08-10 12:53 ` Paolo Bonzini
2017-08-10 14:16 ` Richard W.M. Jones
2017-08-10 14:30 ` Paolo Bonzini [this message]
2017-08-10 15:40 ` Richard W.M. Jones
2017-08-10 16:04 ` Paolo Bonzini
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=367767f5-b3b2-6905-759d-d97196e7f140@redhat.com \
--to=pbonzini@redhat.com \
--cc=hch@lst.de \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-scsi@vger.kernel.org \
--cc=martin.petersen@oracle.com \
--cc=rjones@redhat.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
Powered by JetHome