qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Maxim Levitsky <mlevitsk@redhat.com>
To: Paolo Bonzini <pbonzini@redhat.com>, qemu-devel@nongnu.org
Cc: stefanha@redhat.com
Subject: Re: [PATCH 07/10] scsi/scsi-bus: scsi_device_find: don't return unrealized devices
Date: Wed, 30 Sep 2020 17:32:05 +0300	[thread overview]
Message-ID: <c3dee346c4c949f90566ef1d9f816785d0f175c1.camel@redhat.com> (raw)
In-Reply-To: <20200925172604.2142227-8-pbonzini@redhat.com>

On Fri, 2020-09-25 at 13:26 -0400, Paolo Bonzini wrote:
> From: Maxim Levitsky <mlevitsk@redhat.com>
> 
> The device core first places a device on the bus and then realizes it.
> Make scsi_device_find avoid returing such devices to avoid
> races in drivers that use an iothread (currently virtio-scsi)
> 
> Bugzilla: https://bugzilla.redhat.com/show_bug.cgi?id=1812399
> 
> Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
> Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
> Message-Id: <20200913160259.32145-7-mlevitsk@redhat.com>
> Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
> ---
>  hw/scsi/scsi-bus.c | 83 +++++++++++++++++++++++++++++-----------------
>  1 file changed, 53 insertions(+), 30 deletions(-)
> 
> diff --git a/hw/scsi/scsi-bus.c b/hw/scsi/scsi-bus.c
> index 4ab9811cd8..7599113efe 100644
> --- a/hw/scsi/scsi-bus.c
> +++ b/hw/scsi/scsi-bus.c
> @@ -24,6 +24,55 @@ static void scsi_target_free_buf(SCSIRequest *req);
>  
>  static int next_scsi_bus;
>  
> +static SCSIDevice *do_scsi_device_find(SCSIBus *bus,
> +                                       int channel, int id, int lun,
> +                                       bool include_unrealized)
> +{
> +    BusChild *kid;
> +    SCSIDevice *retval = NULL;
> +
> +    QTAILQ_FOREACH_RCU(kid, &bus->qbus.children, sibling) {
> +        DeviceState *qdev = kid->child;
> +        SCSIDevice *dev = SCSI_DEVICE(qdev);
> +
> +        if (dev->channel == channel && dev->id == id) {
> +            if (dev->lun == lun) {
> +                retval = dev;
> +                break;
Indeed, here no need to use goto, I probably removed some code that needed it
before or just forgot about.
> +            }
> +
> +            /*
> +             * If we don't find exact match (channel/bus/lun),
> +             * we will return the first device which matches channel/bus
> +             */
> +
> +            if (!retval) {
> +                retval = dev;
> +            }
> +        }
> +    }
> +
> +    /*
> +     * This function might run on the IO thread and we might race against
> +     * main thread hot-plugging the device.
> +     * We assume that as soon as .realized is set to true we can let
> +     * the user access the device.
> +     */
> +
> +    if (retval && !include_unrealized &&
> +        !qatomic_load_acquire(&retval->qdev.realized)) {
> +        retval = NULL;
> +    }
> +
> +    return retval;
> +}
I guess you dropped the RCU locking here since you want the callers
of this function to do it. I think it makes lot of sense.

> +
> +SCSIDevice *scsi_device_find(SCSIBus *bus, int channel, int id, int lun)
> +{
> +    RCU_READ_LOCK_GUARD();
> +    return do_scsi_device_find(bus, channel, id, lun, false);
> +}
> +
>  static void scsi_device_realize(SCSIDevice *s, Error **errp)
>  {
>      SCSIDeviceClass *sc = SCSI_DEVICE_GET_CLASS(s);
> @@ -137,7 +186,10 @@ static bool scsi_bus_is_address_free(SCSIBus *bus,
>  				     int channel, int target, int lun,
>  				     SCSIDevice **p_dev)
>  {
> -    SCSIDevice *d = scsi_device_find(bus, channel, target, lun);
> +    SCSIDevice *d;
> +
> +    RCU_READ_LOCK_GUARD();
> +    d = do_scsi_device_find(bus, channel, target, lun, true);
>      if (d && d->lun == lun) {
>          if (p_dev) {
>              *p_dev = d;
> @@ -1570,35 +1622,6 @@ static char *scsibus_get_fw_dev_path(DeviceState *dev)
>                             qdev_fw_name(dev), d->id, d->lun);
>  }
>  
> -SCSIDevice *scsi_device_find(SCSIBus *bus, int channel, int id, int lun)
> -{
> -    BusChild *kid;
> -    SCSIDevice *target_dev = NULL;
> -
> -    RCU_READ_LOCK_GUARD();
> -    QTAILQ_FOREACH_RCU(kid, &bus->qbus.children, sibling) {
> -        DeviceState *qdev = kid->child;
> -        SCSIDevice *dev = SCSI_DEVICE(qdev);
> -
> -        if (dev->channel == channel && dev->id == id) {
> -            if (dev->lun == lun) {
> -                return dev;
> -            }
> -
> -            /*
> -             * If we don't find exact match (channel/bus/lun),
> -             * we will return the first device which matches channel/bus
> -             */
> -
> -            if (!target_dev) {
> -                target_dev = dev;
> -            }
> -        }
> -    }
> -
> -    return target_dev;
> -}
> -
>  /* SCSI request list.  For simplicity, pv points to the whole device */
>  
>  static int put_scsi_requests(QEMUFile *f, void *pv, size_t size,

Best regards,
	Maxim Levitsky



  reply	other threads:[~2020-09-30 14:34 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-25 17:25 [PATCH 00/10] Fix scsi devices plug/unplug races w.r.t virtio-scsi iothread Paolo Bonzini
2020-09-25 17:25 ` [PATCH 01/10] qdev: add "check if address free" callback for buses Paolo Bonzini
2020-09-28  9:30   ` Stefan Hajnoczi
2020-09-30 17:48     ` Paolo Bonzini
2020-09-30 14:27   ` Maxim Levitsky
2020-09-30 23:37     ` Paolo Bonzini
2020-09-25 17:25 ` [PATCH 02/10] scsi: switch to bus->check_address Paolo Bonzini
2020-09-28  9:43   ` Stefan Hajnoczi
2020-09-30 14:28   ` Maxim Levitsky
2020-09-25 17:25 ` [PATCH 03/10] scsi/scsi_bus: switch search direction in scsi_device_find Paolo Bonzini
2020-09-25 17:25 ` [PATCH 04/10] device_core: use drain_call_rcu in in hmp_device_del/qmp_device_add Paolo Bonzini
2020-09-25 17:25 ` [PATCH 05/10] device-core: use RCU for list of children of a bus Paolo Bonzini
2020-09-30 14:29   ` Maxim Levitsky
2020-09-25 17:26 ` [PATCH 06/10] device-core: use atomic_set on .realized property Paolo Bonzini
2020-09-30 14:31   ` Maxim Levitsky
2020-09-30 17:44     ` Paolo Bonzini
2020-09-30 18:03       ` Maxim Levitsky
2020-09-25 17:26 ` [PATCH 07/10] scsi/scsi-bus: scsi_device_find: don't return unrealized devices Paolo Bonzini
2020-09-30 14:32   ` Maxim Levitsky [this message]
2020-09-25 17:26 ` [PATCH 08/10] scsi/scsi_bus: Add scsi_device_get Paolo Bonzini
2020-09-30 14:32   ` Maxim Levitsky
2020-09-30 17:46     ` Paolo Bonzini
2020-09-30 18:04       ` Maxim Levitsky
2020-09-25 17:26 ` [PATCH 09/10] virtio-scsi: use scsi_device_get Paolo Bonzini
2020-09-25 17:26 ` [PATCH 10/10] scsi/scsi_bus: fix races in REPORT LUNS Paolo Bonzini
2020-09-30 14:34   ` Maxim Levitsky
2020-09-25 19:26 ` [PATCH 00/10] Fix scsi devices plug/unplug races w.r.t virtio-scsi iothread no-reply
2020-09-25 22:52 ` no-reply
2020-09-26  0:28 ` no-reply
2020-09-26  0:44 ` no-reply
2020-09-26  1:05 ` no-reply

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=c3dee346c4c949f90566ef1d9f816785d0f175c1.camel@redhat.com \
    --to=mlevitsk@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=stefanha@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;
as well as URLs for NNTP newsgroup(s).