1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
|
[[!meta copyright="Copyright © 2013 Free Software Foundation, Inc."]]
[[!meta license="""[[!toggle id="license" text="GFDL 1.2+"]][[!toggleable
id="license" text="Permission is granted to copy, distribute and/or modify this
document under the terms of the GNU Free Documentation License, Version 1.2 or
any later version published by the Free Software Foundation; with no Invariant
Sections, no Front-Cover Texts, and no Back-Cover Texts. A copy of the license
is included in the section entitled [[GNU Free Documentation
License|/fdl]]."]]"""]]
Some *cloud*y things.
[[!toc]]
# [[!wikipedia OpenStack]]
## IRC, freenode, #hurd, 2013-09-21
<jproulx> Hmmm, was hoping to run hurd on my kvm based openstack cloud, but
no virtio.
<jproulx> I see "Write virtio drivers for KVM. Ideally they would be
userland" is listed as a "small hack", as a sysadmin rather than an OS
hacker it doesn't sound small to me, but if there's some standard
documentation on porting drivers I could take a run at it.
<youpi> well, perhaps "small" is not the proper word
<youpi> compared to e.g. revamping disk i/o :)
<youpi> it's not something one can achieve in e.g. 1h, for instance
<youpi> it's not something straightforward either, one has to get
documentation about virtio (I don't know what exists), and get
documentation about the mach device interface (that's in the gnumach
manual, the devnode translator can be used as a skeleton)
<youpi> jproulx: openstack imposes the use of virtio drivers? that's odd
<jproulx> that's more like I'd expect. I there's enough search terms in
your response for me to see what's really involved
<jproulx> youpi it doesn't impose that but it is how mine is configured the
other thousand VMs are happier that way.
<jproulx> I can look at that side too and see if I need to have everything
use the same device settings or if I can control it per instance
<jproulx> A bit of a non-sequitur at this point but just in case someone
searches the transcripts and sees my questions about hurd on openstack,
yes it is possible to specify non-virtio devices per image, here's the
commandline to load sthibault's qemu image into openstack with devices
that work:
<jproulx> glance image-create --property hw_disk_bus=ide --property
hw_cdrom_bus=ide --property hw_vif_model=rtl8139 --disk-format raw
--container-format bare --name gnu-hurd --copy-from
http://people.debian.org/~sthibault/hurd-i386/debian-hurd.img
<youpi> jproulx: thanks, I've pushed it on the wiki
|