From: Gerd Hoffmann <kraxel@redhat.com>
To: Anthony Liguori <aliguori@us.ibm.com>
Cc: qemu-devel@nongnu.org, Dustin Kirkland <kirkland@canonical.com>
Subject: [Qemu-devel] Re: [PATCH] [STABLE] Fix virtio-blk hot add after remove
Date: Thu, 15 Oct 2009 10:00:01 +0200 [thread overview]
Message-ID: <4AD6D681.3050401@redhat.com> (raw)
In-Reply-To: <1255539554-7956-1-git-send-email-aliguori@us.ibm.com>
[-- Attachment #1: Type: text/plain, Size: 550 bytes --]
On 10/14/09 18:59, Anthony Liguori wrote:
> qdev_init_bdrv() expects that each drive added is the next logical unit for
> the given interface type. However, when dealing with hotplug, there may
> be holes in the units. drive_init reclaims holes in units but qdev_init_bdrv()
> is not smart enough to do this.
Oh, right.
> Suggestions for a less ugly solution are appreciated.
See attached patch. Isn't exactly pretty. But at least it is less
invasive and IMHO the logic is easier to understand as well.
Warning: *untested*.
cheers,
Gerd
[-- Attachment #2: 0001-fix-virtio-blk-hotplugging.patch --]
[-- Type: text/plain, Size: 1891 bytes --]
>From 572ff4912ed021c7b4bb2076d8d72e497f0c434b Mon Sep 17 00:00:00 2001
From: Gerd Hoffmann <kraxel@redhat.com>
Date: Thu, 15 Oct 2009 09:55:11 +0200
Subject: [PATCH] fix virtio blk hotplugging.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
---
hw/pci-hotplug.c | 2 ++
hw/qdev.c | 8 ++++++++
sysemu.h | 1 +
3 files changed, 11 insertions(+), 0 deletions(-)
diff --git a/hw/pci-hotplug.c b/hw/pci-hotplug.c
index 8bedea2..658564e 100644
--- a/hw/pci-hotplug.c
+++ b/hw/pci-hotplug.c
@@ -147,7 +147,9 @@ static PCIDevice *qemu_pci_hot_add_storage(Monitor *mon,
drives_table[drive_idx].unit);
break;
case IF_VIRTIO:
+ drives_table[drive_idx].qdev_pick_this_one_please = 1;
dev = pci_create("virtio-blk-pci", devaddr);
+ drives_table[drive_idx].qdev_pick_this_one_please = 0;
break;
default:
dev = NULL;
diff --git a/hw/qdev.c b/hw/qdev.c
index faecc76..9dd896b 100644
--- a/hw/qdev.c
+++ b/hw/qdev.c
@@ -191,6 +191,14 @@ BlockDriverState *qdev_init_bdrv(DeviceState *dev, BlockInterfaceType type)
int unit = next_block_unit[type]++;
int index;
+ for (index = 0; index < MAX_DRIVES; index++) {
+ if (!drives_table[index].used)
+ continue;
+ if (!drives_table[index].qdev_pick_this_one_please)
+ continue;
+ return drives_table[index].bdrv;
+ }
+
index = drive_get_index(type, 0, unit);
if (index == -1) {
return NULL;
diff --git a/sysemu.h b/sysemu.h
index ce25109..6dc2b4f 100644
--- a/sysemu.h
+++ b/sysemu.h
@@ -173,6 +173,7 @@ typedef struct DriveInfo {
int bus;
int unit;
int used;
+ int qdev_pick_this_one_please; /* band-aid for virtio-blk hotplug */
int drive_opt_idx;
BlockInterfaceErrorAction onerror;
char serial[BLOCK_SERIAL_STRLEN + 1];
--
1.6.2.5
prev parent reply other threads:[~2009-10-15 8:00 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2009-10-14 16:59 [Qemu-devel] [PATCH] [STABLE] Fix virtio-blk hot add after remove Anthony Liguori
2009-10-14 18:22 ` [Qemu-devel] " Dustin Kirkland
2009-10-15 8:00 ` Gerd Hoffmann [this message]
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=4AD6D681.3050401@redhat.com \
--to=kraxel@redhat.com \
--cc=aliguori@us.ibm.com \
--cc=kirkland@canonical.com \
--cc=qemu-devel@nongnu.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
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).