We want to do (at least) two things to the virtio-balloon device:
suppress it, and control its PCI address. Option -no-virtio-balloon
lets us do only the former. To get the latter, replace
-no-virtio-balloon with
Jan Kiszka [Sat, 27 Jun 2009 07:59:40 +0000 (09:59 +0200)]
sdl: Fix memory leakage
Valgrind was so kind to remark that no one bothers to release keycodes
after use and that something is fishy about cleaning up the requested
keyboard descriptor. With this patch applied, we no longer leak about
12k during startup.
Jan Kiszka [Sat, 27 Jun 2009 07:27:22 +0000 (09:27 +0200)]
KVM: x86: Refactor persistent CPU state
This patch aligns the KVM-related layout and encoding of the CPU state
to be saved to disk or migrated with qemu-kvm. The major differences are
reordering of fields and a compressed interrupt_bitmap into a single
number as there can be no more than one pending IRQ at a time.
Jan Kiszka [Thu, 25 Jun 2009 06:23:39 +0000 (08:23 +0200)]
x86: Factor out pc_new_cpu
At this point, this refactoring looks like overkill. But we will need it
for CPU hotplugging, and qemu-kvm already carries it. Merging it early
would help qemu-kvm when rebasing against upstream.
Jan Kiszka [Thu, 25 Jun 2009 06:22:08 +0000 (08:22 +0200)]
monitor: Refactor acl commnds
Refactor the ACL monitor interface to make full use of the monitor
command dispatcher. This also gives proper help formatting and command
completion. Note that 'acl allow' and 'acl deny' were combined to
'acl_add aclname match allow|deny [index]' for consistency reasons.
Jan Kiszka [Sat, 27 Jun 2009 07:25:07 +0000 (09:25 +0200)]
Revert "Introduce reset notifier order"
This reverts commit 8217606e6edb49591b4a6fd5a0d1229cebe470a9 (and
updates later added users of qemu_register_reset), we solved the
problem it originally addressed less invasively.
Jan Kiszka [Sat, 27 Jun 2009 07:24:58 +0000 (09:24 +0200)]
kvm: Rework VCPU synchronization
During startup and after reset we have to synchronize user space to the
in-kernel KVM state. Namely, we need to transfer the VCPU registers when
they change due to VCPU as well as APIC reset.
This patch refactors the required hooks so that kvm_init_vcpu registers
its own per-VCPU reset handler and adds a cpu_synchronize_state to the
APIC reset. That way we no longer depend on the new reset order (and can
drop this disliked interface again) and we can even drop a KVM hook in
main().
Jan Kiszka [Thu, 25 Jun 2009 22:04:00 +0000 (00:04 +0200)]
monitor: Drop pci_addr prefix from hotplug commands
The "pci_addr=" prefix currently required by pci_add/remove and
drive_add has no practical use. Drop it, but still silently accept it
for backward compatibility.
Kevin Wolf [Fri, 26 Jun 2009 18:19:38 +0000 (20:19 +0200)]
qcow2: Cache refcount blocks during snapshot creation
The really time consuming part of snapshotting is to adjust the reference count
of all clusters. Currently after each adjusted cluster the refcount block is
written to disk.
Don't write each single byte immediately to disk but cache all writes to the
refcount block and write them out once we're done with the block.
Kevin Wolf [Fri, 26 Jun 2009 17:51:24 +0000 (19:51 +0200)]
block-raw: Allow pread beyond the end of growable images
When using O_DIRECT, qcow2 snapshots didn't work any more for me. In the
process of creating the snapshot, qcow2 tries to pwrite some new information
(e.g. new L1 table) which will often end up being after the old end of the
image file. Now pwrite tries to align things and reads the old contents of the
file, read returns 0 because there is nothing to read after the end of file and
pwrite is stuck in an endless loop.
This patch allows to pread beyond the end of an image file. Whenever the
given offset is after the end of the image file, the read succeeds and fills
the buffer with zeros.
Alexander Graf [Mon, 29 Jun 2009 13:37:39 +0000 (15:37 +0200)]
Multiboot support v5
This patch implements support for Multiboot on x86 for -kernel.
Multiboot is a "new" approach to get rid of different bootloaders, providing
a unified interface for the kernel. It supports command line options and
kernel modules.
The two probably best known projects using multiboot are Xen and GNU Hurd.
This implementation should be mostly feature-complete. It is missing VBE
extensions, but as no system uses them currently it does not really hurt.
To use multiboot, specify the kernel as -kernel option. Modules should be given
as -initrd options, seperated by a comma (,). -append also works.
Please bear in mind that grub also does gzip decompression, which qemu does
not do yet. To run existing images, please ungzip them first.
The guest multiboot loader code is implemented as option rom using int 19.
Parts of the work are based on efforts by Rene Rebe, who originally ported
my code to int 19.
Andre Przywara [Wed, 24 Jun 2009 22:08:04 +0000 (00:08 +0200)]
preserve the hypervisor bit while KVM trims the CPUID bits
The KVM kernel will disable all bits in CPUID which are not present in
the host. As this is mostly true for the hypervisor bit (1.ecx),
preserve its value before the trim and restore it afterwards.
Andre Przywara [Wed, 24 Jun 2009 22:08:03 +0000 (00:08 +0200)]
remove CPUID host hacks
KVM provides an in-kernel feature to disable CPUID bits that are not
present in the current host. So there is no need here to duplicate this
work. Additionally allows 3DNow! on capable processors, since the
restriction seems to apply to QEMU/TCG only.
Andre Przywara [Wed, 24 Jun 2009 22:08:02 +0000 (00:08 +0200)]
fix KVMs GET_SUPPORTED_CPUID feature usage
If we want to trim the user provided CPUID bits for KVM to be not greater
than that of the host, we should not remove the bits _after_ we sent
them to the kernel.
This fixes the masking of features that are not present on the host by
moving the trim function and it's call from helper.c to kvm.c.
It helps to use -cpu host.
Andre Przywara [Wed, 24 Jun 2009 22:08:00 +0000 (00:08 +0200)]
introduce -cpu host target
Although the guest's CPUID bits can be controlled in a fine grained way
in QEMU, a simple way to inject the host CPU is missing. This is handy
for KVM desktop virtualization, where one wants the guest to support the
full host feature set.
Introduce another CPU type called 'host', which will propagate the host's
CPUID bits to the guest. Unwanted bits can still be turned off by using
the existing syntax (-cpu host,-skinit)
Andre Przywara [Wed, 24 Jun 2009 22:07:59 +0000 (00:07 +0200)]
allow hypervisor CPUID bit to be overriden
KVM defaults to the hypervisor CPUID bit to be set, whereas pure
QEMU clears it. On some occasions one wants to set or clear it the
other way round (for instance to get HyperV running inside a guest).
Move the bit-set to be done before the command line parsing and
enable it by default. One can disable it by using: -cpu qemu64,-hypervisor
Fix some whitespace damage on the way.
Ram Pai [Sat, 27 Jun 2009 00:41:50 +0000 (17:41 -0700)]
support colon in filenames
Problem: It is impossible to feed filenames with the character colon because
qemu interprets such names as a protocol. For example filename scsi:0, is
interpreted as a protocol by name "scsi".
This patch allows user to espace colon characters. For example the above
filename can now be expressed either as 'scsi\:0' or as file:scsi:0
anything following the "file:" tag is interpreted verbatin. However if "file:"
tag is omitted then any colon characters in the string must be escaped using
backslash.
Here are couple of examples:
scsi\:0\:abc is a local file scsi:0:abc
http\://myweb is a local file by name http://myweb
file:scsi:0:abc is a local file scsi:0:abc
file:http://myweb is a local file by name http://myweb
Jan Kiszka [Wed, 24 Jun 2009 10:29:11 +0000 (12:29 +0200)]
Prefer ncurses over curses
Not every distro provides libcurses anymore, at least OpenSUSE, and at
least under a standard library search path. So try to link against
standard ncurses first and then fall back to legacy curses.
Jan Kiszka [Wed, 24 Jun 2009 12:42:32 +0000 (14:42 +0200)]
slirp: Make hostfwd_add/remove multi-instance-aware
Extend the syntax of hostfwd_add/remove to optionally take a tuple of
VLAN ID and slirp stack name. If those are omitted, the commands will
continue to work on the first registered slirp stack.
Jan Kiszka [Wed, 24 Jun 2009 12:42:31 +0000 (14:42 +0200)]
slirp: Enable multi-instance support for the smb service
Push the smb state, smb_dir, into SlirpState and construct it in a way
that allows multiple smb instances (one per slirp stack). Remove the smb
directory on slirp cleanup instead of qemu termination. As VLAN clients
are also cleaned up on process termination, no feature is lost.
Jan Kiszka [Wed, 24 Jun 2009 12:42:31 +0000 (14:42 +0200)]
slirp: Enable multiple instances
Once again this was a long journey to reach the destination: Allow to
instantiate slirp multiple times. But as in the past, the journey was
worthwhile, cleaning up, fixing and enhancing various parts of the user
space network stack along the way.
What is this particular change good for? Multiple slirps instances
allow separated user space networks for guests with multiple NICs. This
is already possible, but without any slirp support for the second
network, ie. without a chance to talk to that network from the host via
IP. We have a legacy guest system here that benefits from this slirp
enhancement, allowing us to run both of its NICs purely over
unprivileged user space IP stacks.
Another benefit of this patch is that it simply removes an artificial
restriction of the configuration space qemu is providing, avoiding
another source of surprises that users may face when playing with
possible setups.
Jan Kiszka [Wed, 24 Jun 2009 12:42:31 +0000 (14:42 +0200)]
slirp: Allocate/free stack instance dynamically
Allocate the internal slirp state dynamically and provide and call
slirp_cleanup to properly release it after use. This patch finally
unbreaks slirp release and re-instantiation via host_net_* monitor
commands.
Jan Kiszka [Wed, 24 Jun 2009 12:42:31 +0000 (14:42 +0200)]
slirp: Use internal state in interface
This now also exports the internal state to the slirp users in qemu,
returning it from slirp_init and expecting it along with service
invocations. Additionally provide an opaque value interface for the
callbacks from slirp into the qemu core.
Jan Kiszka [Wed, 24 Jun 2009 12:42:31 +0000 (14:42 +0200)]
slirp: Factor out internal state structure
The essence of this patch is to stuff (almost) all global variables of
the slirp stack into the structure Slirp. In this step, we still keep
the structure as global variable, directly accessible by the whole
stack. Changes to the external interface of slirp will be applied in
the following patches.
Jan Kiszka [Wed, 24 Jun 2009 12:42:30 +0000 (14:42 +0200)]
slirp: Kill slirp_is_inited
Avoid the need for slirp_is_inited by refactoring the protected
slirp_select_* functions. This also avoids the clearing of all fd sets
on select errors.
Jan Kiszka [Wed, 24 Jun 2009 12:42:30 +0000 (14:42 +0200)]
slirp: Make IP packet ID consistent
Currently, ip_id is always initialized to 0 on slirp startup (despite
the broken attempt to derive it from the clock). This is good for
reproducibility. But it is not preserved across save/restore. This patch
therefore drops the dead initialization code from ip_init and introduces
ip_id to the persistent slirp state.
Jan Kiszka [Wed, 24 Jun 2009 12:42:30 +0000 (14:42 +0200)]
slirp: Factor out one-time initialization
In order to prepare re-initialization and multi-instance slirp, factor
out init code that is of global scope and (at least for now) only need
to be run once.
This also fixes the potentially uninitialized use of our_addr in
get_dns_addr.
Jan Kiszka [Wed, 24 Jun 2009 12:42:30 +0000 (14:42 +0200)]
slirp: tftp: Rework filename handling
This changes the filename handling from a static buffer in tftp_session
for the client-provided name + prefix to a dynamically allocated buffer
that keeps the combined path in one place.
Jan Kiszka [Wed, 24 Jun 2009 12:42:29 +0000 (14:42 +0200)]
slirp: Drop dead code
After all its years inside the qemu tree, there is no point in keeping
the dead code paths of slirp. This patch is a first round of removing
usually commented out code parts. More cleanups need to follow (and
maybe finally a proper reindention).
Jan Kiszka [Wed, 24 Jun 2009 12:42:29 +0000 (14:42 +0200)]
slirp: Add info usernet for dumping connection states
Break out sockstats from the slirp statistics and present them under the
new info category "usernet". This patch also improves the current output
/wrt proper reporting connection source and destination.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
slirp: Bind support for host forwarding rules
Extend the hostfwd rule format so that the user can specify on which
host interface qemu should listen for incoming connections. If omitted,
binding will takes place against all interfaces.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
slirp: Rework monitor commands for host forwarding
Improve the monitor interface for adding and removing host forwarding
rules by splitting it up in two commands and rename them to hostfwd_add
and hostfwd_remove. Also split up the paths taken for legacy -redir
support and the monitor add command as the latter will be extended later
on.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
slirp: Fix port comparision in slirp_remove_hostfwd
For UDP host forwardings, fport is not stable, every outgoing packet of
the redirection can modify it. Use getsockname instead to look up the
port that is actually used on the host side.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
slirp: Rework external configuration interface
With the internal IP configuration made more flexible, we can now
enhance the user interface. This patch adds a number of new options to
"-net user": net (address and mask), host, dhcpstart, dns and smbserver.
It also renames "redir" to "hostfwd" and "channel" to "guestfwd" in
order to (hopefully) clarify their meanings. The format of guestfwd is
extended so that the user can define not only the port but also the
virtual server's IP address the forwarding starts from.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
slirp: Rework internal configuration
The user mode IP stack is currently only minimally configurable /wrt to
its virtual IP addresses. This is unfortunate if some guest has a fixed
idea of which IP addresses to use.
Therefore this patch prepares the stack for fully configurable IP
addresses and masks. The user interface and default addresses remain
untouched in this step, they will be enhanced in the following patch.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
slirp: Move smb, redir, tftp and bootp parameters and -net channel
So far a couple of slirp-related parameters were expressed via
stand-alone command line options. This it inconsistent and unintuitive.
Moreover, it prevents both dynamically reconfigured (host_net_add/
delete) and multi-instance slirp.
This patch refactors the configuration by turning -smb, -redir, -tftp
and -bootp as well as -net channel into options of "-net user". The old
stand-alone command line options are still processed, but no longer
advertised. This allows smooth migration of management applications to
to the new syntax and also the extension of that syntax later in this
series.
Jan Kiszka [Wed, 24 Jun 2009 12:42:28 +0000 (14:42 +0200)]
Introduce get_next_param_value
In order to parse multiple instances of the same param=value pair,
introduce get_next_param_value which can pass back to string parsing
position after reading a parameter value.
Hi all,
this patch implements zooming capabilities for the sdl interface.
A new sdl_zoom_blit function is added that is able to scale and blit a
portion of a surface into another.
This way we can enable SDL_RESIZABLE and have a real_screen surface with
a different size than the guest surface and let sdl_zoom_blit take care
of the problem.
john cooper [Mon, 22 Jun 2009 18:26:51 +0000 (14:26 -0400)]
Add serial number support for virtio_blk
[brought forward to current qemu-kvm.git]
This patch implements the missing qemu logic to
interpret a '-drive .. serial=XYZ ..' flag for
a virtio_blk device.
The serial number string is contained in a
skeletal IDENTIFY DEVICE data structure and
this structure is made available to the guest
virtio_blk driver via pci i/o region 0.
Add commands to exercise asynchronous reads/writes and to flush all
outstanding aio commands. Commands to exercise aio cancellations will
follow in a separate patch.
Stefan Weil [Sun, 21 Jun 2009 16:35:03 +0000 (18:35 +0200)]
Fix dump output in qemu-io.
The dump output was not nicely formatted for bytes
larger than 0x7f, because signed values expanded to
sizeof(int) bytes. So for example 0xab did not print
as "ab", but as "ffffffab".
I also cleaned the function prototype, which avoids
new type casts and allows to remove an existing
type cast.
qemu/net: flag to control the number of vectors a nic has
Add an option to specify the number of MSI-X vectors for PCI NIC cards. This
can also be used to disable MSI-X, for compatibility with old qemu. This
option currently only affects virtio cards.
qemu/virtio: virtio support for many interrupt vectors
Extend virtio to support many interrupt vectors, and rearrange code in
preparation for multi-vector support (mostly move reset out to bindings,
because we will have to reset the vectors in transport-specific code).
Actual bindings in pci, and use in net, to follow.
Load and save are not connected to bindings yet, so they are left
stubbed out for now.
qemu/apic: minimal MSI/MSI-X implementation for PC
Implement MSI support in APIC. Note that MSI and MMIO APIC registers
are at the same memory location, but actually not on the global bus: MSI
is on PCI bus, APIC is connected directly to the CPU. We map them on the
global bus at the same address which happens to work because MSI
registers are reserved in APIC MMIO and vice versa.
Add functions implementing MSI-X support. First user will be virtio-pci.
Note that platform must set a flag to declare MSI supported: this
is a safety measure to avoid breaking platforms which should support
MSI-X but currently lack this in the interrupt controller emulation.
For PC this will be set by APIC.
projects / qemu.git / log