Like qdev_init(), but terminate program via hw_error() instead of
returning an error value.
Use it instead of qdev_init() where terminating the program on failure
is okay, either because it's during machine construction, or because
we know that failure can't happen.
Because relying in the latter is somewhat unclean, and the former is
not always obvious, it would be nice to go back to qdev_init() in the
not-so-obvious cases, only with proper error handling. I'm leaving
that for another day, because it involves making sure that error
values are properly checked by all callers.
Patchworks-ID: 35168
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
But do so only where it may actually fail. Leave the rest for the
next commit.
Patchworks-ID: 35167
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Making pci device cleanup msix automatically makes pci.c depend on
msix.c, which is IMO messy. Since devices do msix_init it's easy and
natural for them to also do msix_uninit.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
PCI load routine has to be called with size equal to 256 (otherwise it
will crash in weird ways). So assert this, making code clearer.
Also avoid dynamically sized array on stack - good for portability.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Cc: Juan Quintela <quintela@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
It's qdev_create() specialized for PCI, so name it accordingly.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Monitor command "pci_add ADDR nic model=MODEL" uses pci_nic_init() to
create the NIC. When MODEL is unknown or "?", this prints to stderr
and terminates the program.
Change pci_nic_init() not to treat "?" specially, and to return NULL
on failure. Switch uses during startup to new convenience wrapper
pci_nic_init_nofail(), which behaves just like pci_nic_init() used to
do.
Bonus bug fix: we now check for qdev_init() failing there.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Before this patch, pci_nic_init() returns NULL when it can't find the
model in pci_nic_models[]. Except this can't happen, because
qemu_check_nic_model_list() just searched for model in
pci_nic_models[], and terminated the program on failure.
Repeating the search here is pointless. Instead, change
qemu_check_nic_model_list() to return the model's array index.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Monitor command "pci_add ADDR storage ..." does its work in
qemu_pci_hot_add_nic(). It called pci_create(..., ADDR) to create the
device. That's wrong, because pci_create() terminates the program
when ADDR is invalid.
Use pci_get_bus_devfn() and pci_create_noinit() instead.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Reset BARs and a couple of other registers on bus reset, as per PCI
spec.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
refactor code slightly, adding symbolic constants and functions, and
using macros where possible. This will also make following reset
patches easier.
No functional changes.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Create qdev infrastructure for pci hotplug. PCI bus implementations
must register a handler for hotplug. Creating a new PCI device will
automagically hot-plug it in case the PCI bus in question supports this.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
pci_unregister_device is static now and hooked into Devicestate->exit.
qdev_free(pci_device) works now.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
If the user issues one of the following commands to the Monitor:
pci_add pci_addr=auto nic model=None
pci_add pci_addr=auto nic model=?
QEMU will exit, because the function used to perform sanity
checks (qemu_check_nic_model_list()) exits on error.
This function is used by the startup code, where it makes
sense to exit on error, but in the Monitor it doesn't.
Changing qemu_check_nic_model_list() to not exit on error
is not possible though, as it's used by the board init
code (the PC one), where all board specific code must have
void return.
The way I've chosen to fix this was to introduce a new function
called pci_nic_supported(), which checks if the NIC is supported
and returns true or false accordingly.
The new function is used only by the Monitor, it performs the
necessary check and returns an error in case the NIC is not
supported, thus qemu_check_nic_model_list()'s exit is never trigged.
The following should be observed:
1. Only the specified NIC is checked, the default one is assumed
to be supported
2. The NIC query command (model=?) won't work with pci_add, the
right way to do this with the Monitor is to add a new command
Signed-off-by: Luiz Capitulino <lcapitulino@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
In a later patch, we introduce pre_save() and post_save() functions.
The whole point of that operation is to change things in the state.
Without this patch, we have to remove the const qualifier in each
use with a cast
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
In the very least, a change like this requires discussion on the list.
The naming convention is goofy and it causes a massive merge problem. Something
like this _must_ be presented on the list first so people can provide input
and cope with it.
This reverts commit 99a0949b72.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Like pci_create_simple() but doesn't call qdev_init(), so one can
set properties before initializing the device.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Sorry folks, but it has to be. One more of these invasive qdev patches.
We have a serious design bug in the qdev interface: device init
callbacks can't signal failure because the init() callback has no
return value. This patch fixes it.
We have already one case in-tree where this is needed:
Try -device virtio-blk-pci (without drive= specified) and watch qemu
segfault. This patch fixes it.
With usb+scsi being converted to qdev we'll get more devices where the
init callback can fail for various reasons.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
This uses a variant of buffer, with extra checks. Also uses the new
support for cheking that a read value is less or equal than a field.
Signed-off-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
This patch is a major overhaul of the device properties. The properties
are saved directly in the device state struct now, the linked list of
property values is gone.
Advantages:
* We don't have to maintain the list with the property values.
* The value in the property list and the value actually used by
the device can't go out of sync any more (used to happen for
the pci.devfn == -1 case) because there is only one place where
the value is stored.
* A record describing the property is required now, you can't set
random properties any more.
There are bus-specific and device-specific properties. The former
should be used for properties common to all bus drivers. Typical
use case is bus addressing, i.e. pci.devfn and i2c.address.
Properties have a PropertyInfo struct attached with name, size and
function pointers to parse and print properties. A few common property
types have PropertyInfos defined in qdev-properties.c. Drivers are free
to implement their own very special property parsers if needed.
Properties can have default values. If unset they are zero-filled.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
* fix secondary bus setup.
* use base->name instead of "FIXME" for device name.
Yes, the device name is redundant. Only for drivers converted
to qdev already though. Once all drivers are converted we can
and should kill it.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
BusInfo is filled with name and size (pretty much like I did for
DeviceInfo as well). There is also a function pointer to print
bus-specific device information to the monitor. sysbus is hooked
up there, I've also added a print function for PCI.
Device creation is slightly modified as well: The device type search
loop now also checks the bus type while scanning the list instead of
complaining thereafter in case of a mismatch. This effectively gives
each bus a private namespace for device names.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Paul Brook <paul@codesourcery.com>
Makes pci_qdev_register take a PCIDeviceInfo struct instead of a bunch
of parameters. Also adds config_read and config_write callbacks to
PCIDeviceInfo, so drivers needing these can be converted to the qdev
device API too.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
This reverts commit 8217606e6e (and
updates later added users of qemu_register_reset), we solved the
problem it originally addressed less invasively.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
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.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Add "cmask" table of constant register masks: if a bit is not writeable
and is set in cmask table, this bit is checked on load. An attempt to
load an image that would change such a register causes load to fail.
Use this table to make sure that load does not modify registers that
guest can not change (directly or indirectly).
Note: we can't just assume that read-only registers never change,
because the guest could change a register indirectly.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Add routines to manage PCI capability list. First user will be MSI-X.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Change much of hw/pci to use symbolic constants and a table-driven
design: add a mask table with writable bits set and readonly bits unset.
Detect change by comparing original and new registers.
This makes it easy to support capabilities where read-only/writeable
bit layout differs between devices, depending on capabilities present.
As a result, writing a single byte in BAR registers now works as
it should. Writing to upper limit registers in the bridge
also works as it should. Code is also shorter.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Simply pass the PCI address through qemu_pci_hot_add_nic() to
pci_nic_init() and through qemu_pci_hot_add_storage() to pci_create().
Before, pci_device_hot_add() passed along the PCI bus to use, and
ignored any user-specified slot.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Make net_client_init() accept addr=, put the value into struct
NICinfo. Use it in pci_nic_init(), and remove arguments bus and
devfn.
Don't support addr= in third argument of monitor command pci_add,
because that clashes with its first argument. Admittedly unelegant.
Machines "malta" and "r2d" have a default NIC with a well-known PCI
address. Deal with that the same way as the NIC model: make
pci_nic_init() take an optional default to be used when the user
doesn't specify one.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
This function is used to manage a PCI BAR, so make the more generic
pci_register_io_region() available to other uses.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Rationale: move device information from code to data structures.
v2: Adapt the drivers missed in the first version.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Implement and use a common device bus state. The main side-effect is
that creating a bus and attaching it to a parent device are no longer
separate operations. For legacy code we allow a NULL parent, but that
should go away eventually.
Also tweak creation code to veriry theat a device in on the right bus.
Signed-off-by: Paul Brook <paul@codesourcery.com>
From the documentation I can find, this register is supposed to be read-only.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@7070 c046a42c-6fe2-441c-8c8c-71466251a162
Refactor the monitor API and prepare it for decoupled terminals:
term_print functions are renamed to monitor_* and all monitor services
gain a new parameter (mon) that will once refer to the monitor instance
the output is supposed to appear on. However, the argument remains
unused for now. All monitor command callbacks are also extended by a mon
parameter so that command handlers are able to pass an appropriate
reference to monitor output services.
For the case that monitor outputs so far happen without clearly
identifiable context, the global variable cur_mon is introduced that
shall once provide a pointer either to the current active monitor (while
processing commands) or to the default one. On the mid or long term,
those use case will be obsoleted so that this variable can be removed
again.
Due to the broad usage of the monitor interface, this patch mostly deals
with converting users of the monitor API. A few of them are already
extended to pass 'mon' from the command handler further down to internal
functions that invoke monitor_printf.
At this chance, monitor-related prototypes are moved from console.h to
a new monitor.h. The same is done for the readline API.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6711 c046a42c-6fe2-441c-8c8c-71466251a162
This code parses full PCI device addresses. It then rejects domains
other than zero, because these are not supported in QEMU.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6609 c046a42c-6fe2-441c-8c8c-71466251a162
Otherwise the PCI size for such regions can be calculated erroneously.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6604 c046a42c-6fe2-441c-8c8c-71466251a162
Unregister the pci device, unassign its IO and memory regions, and free
associated data.
Add a callback so drivers can free device state.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6603 c046a42c-6fe2-441c-8c8c-71466251a162
Change the PCI network drivers init functions to return the PCIDev, to
inform which slot has been hot-plugged.
Also record PCIDevice structure on NICInfo to locate for release on
hot-removal.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6593 c046a42c-6fe2-441c-8c8c-71466251a162
Add pci_find_bus/pci_find_device to be used by PCI hotplug.
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6592 c046a42c-6fe2-441c-8c8c-71466251a162
This patch adds and uses #defines for PCI device classes and subclases,
using a new pci_config_set_class() function, similar to the recently
added pci_config_set_vendor_id() and pci_config_set_device_id().
Change since v1: fixed compilation of hw/sun4u.c
Signed-off-by: Stuart Brady <stuart.brady@gmail.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6491 c046a42c-6fe2-441c-8c8c-71466251a162
Add a table of PCI NIC models to pass to qemu_setup_nic_model().
While we're at it, also add a corresponding table of NIC init
functions.
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6287 c046a42c-6fe2-441c-8c8c-71466251a162
The Command register in the PCI config space has some read-only bits.
Any writes to those bits should be masked out.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6092 c046a42c-6fe2-441c-8c8c-71466251a162
The Status register in the PCI config space has some read-only bits.
Any writes to those bits should be masked out.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6091 c046a42c-6fe2-441c-8c8c-71466251a162
This adds virtio-net support. This is based on the virtio-net driver
that exists in kvm-userspace. This also adds a new qemu_sendv_packet
which virtio-net requires.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6073 c046a42c-6fe2-441c-8c8c-71466251a162
This sets a default PCI subsystem ID for all emulated PCI devices. PCI
specs require this, so do it.
In many cases it is enougth to know the PCI ID to handle a device
correctly. Sometimes a device driver must identify the exact piece of
hardware (via PCI Subsystem ID) though.
What does this patch to qemu devices:
Right now the emulated PCI devices have no PCI subsystem ID, only the
PCI ID. The discussed patch sets a default PCI subsystem ID for all
emulated devices. Which will make the qemu devices look pretty much
like in the laptop case: all PCI subsystem IDs will point to qemu by
default.
If a driver emulates a very specific piece of hardware where it has to
emulate more than just the PCI chip, it can overwrite the PCI subsystem
ID without problems. The es1370 driver does that for example.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@5986 c046a42c-6fe2-441c-8c8c-71466251a162
MMIO exits are more expensive in KVM or Xen than in QEMU because they
involve, at least, privilege transitions. However, MMIO write
operations can be effectively batched if those writes do not have side
effects.
Good examples of this include VGA pixel operations when in a planar
mode. As it turns out, we can get a nice boost in other areas too.
Laurent mentioned a 9.7% performance boost in iperf with the coalesced
MMIO changes for the e1000 when he originally posted this work for KVM.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@5961 c046a42c-6fe2-441c-8c8c-71466251a162
Applied %s/^\([^I ]*\)^I/\1 /g on e1000.c and added e1000 to help message.
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3949 c046a42c-6fe2-441c-8c8c-71466251a162
