1. Add correct definitions of error numbers.
2. Implement SYS_osf_sigprocmask
3. Implement SYS_osf_get/setsysinfo for IEEE_FP_CONTROL.
This last requires exposing the FPCR value to do_syscall.
Since this value is actually split up into the float_status,
expose routines from helper.c to access it.
Finally, also add a float_exception_mask field to float_status.
We don't actually use it to control delivery of exceptions to
the emulator yet, but simply hold the value that we placed there
when loading/storing the FPCR.
Nathan Froyd [Tue, 8 Dec 2009 16:06:29 +0000 (08:06 -0800)]
target-mips: add mips16 instruction decoding
There's no good way to add this incrementally, so we do it all at once.
The only changes to shared code are in handle_delay_slot. We need to
flip ISAMode when doing a jump-and-exchange. We also need to set
ISAMode the low bit of the target address for jump-to-register.
Also, since we're now adding bits that can be in MIPS_HFLAG_BMASK_EXT,
make sure we use MIPS_HFLAG_BMASK_BASE in the places where we just want
basic information about a branch.
Nathan Froyd [Tue, 8 Dec 2009 16:06:24 +0000 (08:06 -0800)]
target-mips: move ROTR and ROTRV inside gen_shift_{imm, }
It's easier to implement mips16 shift instructions if we're not
examining the opcode inside gen_shift_{imm,}. So move ROTR and ROTRV
and do the special-case handling of SRL and SRLV inside decode_opc.
Likewise for their 64-bit counterparts.
Nathan Froyd [Tue, 8 Dec 2009 16:06:22 +0000 (08:06 -0800)]
target-mips: add new HFLAGs for JALX and 16/32-bit delay slots
We create separate masks for the "basic" branch hflags and the
"extended" branch hflags and define MIPS_HFLAG_BMASK as the logical or
of those two. This is done to avoid churning the codebase in lots of
different places.
We also make the execution mode an hflag under MIPS_HFLAG_TMASK
workaround for cmd646 bmdma register access while no dma is active
This is a workaround only, and is a partial revert
of a few changes to BMDMAState which removed pci_dev
field on the way.
- cmd646 pci_from_bm() expects bm->unit value to
correspond with bm data being passed to callback
as opaque pointer. This breaks when write to dma
control register of second channel happens when no
dma operation is in progress, so bm->unit is zero
for second channel, and pci_from_bm() returns garbage
pointer. Crash happens shortly after that while
dereferencing that pointer.
s390 code has an obvious typo, which results in:
hw/s390-virtio.c: At top level:
hw/s390-virtio.c:249: error: request for member ‘no_vga’ in something not a structure or union
Building on 32 bit host we get:
hw/s390-virtio.c: In function ‘s390_init’:
hw/s390-virtio.c:184: error: integer constant is too large for ‘unsigned long’ type
64 bit values must be ULL.
Artyom Tarasenko [Sun, 13 Dec 2009 13:30:44 +0000 (13:30 +0000)]
fdc/sparc32: don't hang on detection under OBP
Stepping through the SS-5's OBP initialization routines
it looks like reading fdc main status register should
clear the fd interrupt.
The patch doesn't fix problems with fdc on sparc platform,
it only fixes fdc detection.
Artyom Tarasenko [Sun, 13 Dec 2009 13:30:19 +0000 (13:30 +0000)]
Sparc32: implement AFX for SS-5
Implement a stub for the AFX register on SparcStation-5.
This stub is needed for running the original SS-5 OBP
instead of OpenBIOS (which allows to boot Solaris 2.5.1
and Solaris 2.6 kernels).
Artyom Tarasenko [Sun, 13 Dec 2009 10:32:36 +0000 (11:32 +0100)]
scsi-disk: Inquiry with allocation length of CDB < 36 (v4)
According to the SCSI-2 specification,
http://ldkelley.com/SCSI2/SCSI2/SCSI2/SCSI2-08.html#8.2.5 ,
"if the allocation length of the command descriptor block (CDB) is too
small to transfer all of the parameters, the additional length shall
not be adjusted to reflect the truncation."
The 36 mandatory bytes of response are written to outbuf, and then
only the length requested in CDB is transferred.
Glauber Costa [Thu, 22 Oct 2009 12:26:56 +0000 (10:26 -0200)]
v2: properly save kvm system time msr registers
Currently, the msrs involved in setting up pvclock are not saved over
migration and/or save/restore. This patch puts their value in special
fields in our CPUState, and deal with them using vmstate.
kvm also has to account for it, by including them in the msr list
for the ioctls.
Luiz Capitulino [Thu, 10 Dec 2009 19:16:10 +0000 (17:16 -0200)]
VNC: Convert do_info_vnc() to QObject
Return a QDict with server information. Connected clients are returned
as a QList of QDicts.
The new functions (vnc_qdict_remote_addr(), vnc_qdict_local_addr() and
put_addr_qdict()) are used to insert 'host' and 'service' information
in the returned QDict.
This patch is big, but I don't see how to split it.
Luiz Capitulino [Thu, 10 Dec 2009 19:16:05 +0000 (17:16 -0200)]
migration: Convert do_info_migrate() to QObject
Return a QDict, which may contain up to more two QDicts, depending
on the type of migration we're performing.
IMPORTANT: as a QInt stores a int64_t integer, RAM values are going
to be stored as int64_t and not as uint64_t as they are today. If
this is a problem QInt will have to be changed.
Luiz Capitulino [Thu, 10 Dec 2009 19:15:58 +0000 (17:15 -0200)]
monitor: do_info_version(): Use QDict
All 'info' commands should use QDict, this commit also kills
monitor_print_qobject() as do_info_version() doesn't use it
anymore (and no handler will).
Gerd Hoffmann [Fri, 11 Dec 2009 10:25:07 +0000 (11:25 +0100)]
vnc: fix capslock tracking logic.
The capslock tracking logic added by commit 6b1325029d80455b9da7cd7bd84a88cb915b867c doesn't work correctly for vnc
clients without EXT_KEY_EVENT support. The reason is that qemu converts
keysyms for letters to lowercase for the keysym2scancode lookup. It
then also passes the lowercase value down to do_key_event(), but the
capslock tracking code needs it with the correct case to work properly.
This patch adds a new variable for the lowercase keysym so we'll keep
the unmodified value for do_key_event().
The keysym2scancode is not needed with EXT_KEY_EVENT capable clients
like any app based on the gtk-vnc widget, so I missed that case in
testing ...
Gerd Hoffmann [Thu, 10 Dec 2009 10:11:07 +0000 (11:11 +0100)]
scsi: fix drive hotplug.
This patch fills the DriveInfo->unit after hotplugging a scsi disk.
It makes a difference when auto-assigning a scsi id, where unit was
left filled with '-1' instead of the actual scsi id.
With this patch applied the the drive naming logic in drive_init() works
as good as it did in previous releases. Which means it works fine with
a single scsi bus.
Gerd Hoffmann [Thu, 10 Dec 2009 10:11:06 +0000 (11:11 +0100)]
pci: don't hw_error() when no slot is available.
Current PCI code will simply hw_error() and thus abort in case no free
PCI slot is available or the requested PCI slot is already in use by
another device. For the hotplug case this behavior is not acceptable.
This patch makes qemu pass up the error properly, so the calling code
can decide whenever it wants to exit with an error (on startup) or
whenever it wants to continue (hotplug).
Gerd Hoffmann [Thu, 10 Dec 2009 10:11:05 +0000 (11:11 +0100)]
pci: don't abort() when trying to hotplug with acpi off.
The PCI bus on x86 requires ACPI for hotplug support, thus disbling ACPI
also disables hotplug for the PCI bus. This patch makes qemu check
whenever the PCI bus in question can handle hotplug before trying to add
devices. This is needed because qdev will abort() on any attempt to
hotplug devices into a non-hotpluggable bus.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:54 +0000 (13:11 +0100)]
Set default console to virtio on S390x
All "normal" system emulation targets in qemu I'm aware of display
output on either VGA or serial output.
Our S390x virtio machine doesn't have such kind of legacy hardware. So
instead we need to default to a virtio console.
Add flags to QEMUMachine to indicate which kind of default devices make
sense for the machine in question. Use it for S390x: enable virtcon,
disable serial, parallel and vga.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:53 +0000 (13:11 +0100)]
default devices: virtio consoles.
This patch adds a variable default_virtcon which says whenever a default
virtio console should be added. It is disabled by default, followup
patch will enable it for s390. It is cleared when qemu finds
'-virtiocon', '-device virtio-console-s390' or '-device
virtio-console-pci' on the command line.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:50 +0000 (13:11 +0100)]
rework -monitor handling, switch to QemuOpts
This patch reworks the -monitor handling:
- It adds a new "mon" QemuOpts list for the monitor(s).
- It adds a monitor_parse() function to parse the -monitor switch.
- It adds a mon_init function to initialize the monitor(s) from the
"mon" QemuOpts list.
- It winds up everything and removes the old bits.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:48 +0000 (13:11 +0100)]
default devices: drives
Add a default_drive variable which specified whenever the default drives
(cdrom, floppy, sd) should be created. It is cleared when the new
-nodefaults switch is specified on the command line.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:47 +0000 (13:11 +0100)]
default devices: network
Add a default_net variable which specified whenever a default network
should be created. It is cleared in case any -net option is specified
and it is also added to the new -nodefaults switch.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:45 +0000 (13:11 +0100)]
default devices: vga adapter.
Qemu creates a vga display for you in case you didn't specify one on the
command line. Right now this is tied to the '-vga <type>' command line
switch, which in turn causes trouble if you are creating your gfx card
using '-device VGA,<props>'.
This patch adds a variable default_vga which says whenever a default
serial line should be added. It is enabled by default. It is cleared
when qemu finds '-vga' or '-device {VGA,Cirrus VGA,QEMUware SVGA}' on
the command line.
'-device VGA' still doesn't work though due to a initialization order
issue (vga must init before calling i440fx_init_memory_mappings).
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:43 +0000 (13:11 +0100)]
default devices: qemu monitor.
This patch makes the monitor default device configuration work like the
default serial and parallel port devices. It adds a variable
default_monitor which says whenever a default monitor should be added.
It is enabled by default. It is cleared when qemu finds '-monitor' on
the command line.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:42 +0000 (13:11 +0100)]
default devices: parallel port.
Qemu creates a default parallel port for you in case you didn't specify
one on the command line. Right now this is tied to the '-parallel
<chardev>' command line switch, which in turn causes trouble if you are
creating your parallel port via '-device isa-parallel,<props>'.
This patch adds a variable default_parallel which says whenever a default
parallel port should be added. It is enabled by default. It is cleared
when qemu finds '-parallel' or '-device isa-parallel' on the command line.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:41 +0000 (13:11 +0100)]
default devices: core code & serial lines.
Qemu creates a default serial line for you in case you didn't specify
one on the command line. Right now this is tied to the '-serial
<chardev>' command line switch, which in turn causes trouble if you are
creating your serial line via '-device isa-serial,<props>'.
This patch adds a variable default_serial which says whenever a default
serial line should be added. It is enabled by default. It is cleared
when qemu finds '-serial' or '-device isa-serial' on the command line.
Part of the patch is some infrastructure for the '-device $driver'
checking (default_driver_check function) which will also be used by the
other patches of this series.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:34 +0000 (13:11 +0100)]
qdev: add command line option to set global defaults for properties.
This patch adds infrastructure and command line option for setting
global defaults for device properties, i.e. you can for example use
-global virtio-blk-pci.vectors=0
to turn off msi by default for all virtio block devices. The config
file syntax is:
[global]
driver = "virtio-blk-pci"
property = "vectors"
value = "0"
This can also be used to set properties for devices which are not
created via -device but implicitly via machine init, i.e.
-global isa-fdc,driveA=<name>
This patch uses the mechanism which configures properties for the
compatibility machine types (pc-0.10 & friends). The command line
takes precedence over the machine type values.
Gerd Hoffmann [Tue, 8 Dec 2009 12:11:33 +0000 (13:11 +0100)]
qdev: make compat stuff more generic
This patch renames the compat properties into global properties and
makes them more generic. The compatibility stuff is only one of
multiple possible users now.
Commits c7c338c4, 41471a23, 7a046f5f and a488be27 used
lower_case_with_underscores for class values. Existing usage
CamelCase. ChangeToThatForConsistency.
Device names with whitespace require quoting in the shell and in the
monitor. Some of the offenders are also overly long. Some have a
more convenient alias, some don't.
The place for verbose device names is DeviceInfo member desc. The
name should be short & sweet.
qdev: Separate USB product description from qdev name
Using the qdev name for the product description makes for inconvenient
qdev names.
Put the product description in new USBDeviceInfo member product_desc.
Make usb_qdev_init() use it. No user or guest visible change, since
the value is still the same.
Gleb Natapov [Tue, 8 Dec 2009 13:50:54 +0000 (15:50 +0200)]
fix rtc-td-hack on host without high-res timers
On hosts without high-res timers it is impossible to inject rtc interrupt
faster then 1kHz. Windows sometimes configures RTC to generate 1kHz
interrupts, so we can't inject missed interrupts when running on such
hosts. Always injecting an interrupt on REG_C read is also not an option
since Windows wait for REG_C to become zero with interrupt disabled
during boot. This patch uses mixed approach: accelerate timer + inject
up to 1000 interrupts on REG_C read.
migrating between hosts which have different features
might break silently, if the migration destination
does not support some features supported by source.
Prevent this from happening by comparing acked feature
bits with the mask supported by the device.
Anthony Liguori [Wed, 9 Dec 2009 18:59:36 +0000 (12:59 -0600)]
Do not abort on qemu_malloc(0) in production builds
qemu_malloc() does not allow size=0 to be passed in and aborts on this behavior.
Unfortunately, there is good reason to believe that within qemu, there are a
number of, so far, undetected places that assume size=0 can be safely passed.
Since we do not want to abort unnecessarily in production builds, return
qemu_malloc(1) whenever the version file indicates that this is a production
build.
projects / qemu.git / log