Michael Roth [Tue, 19 Jul 2011 19:50:34 +0000 (14:50 -0500)]
qapi: add QMP output visitor
Type of Visiter class that serves as the inverse of the input visitor:
it takes a series of native C types and uses their values to construct a
corresponding QObject. The command marshaling/dispatcher functions will
use this to convert the output of QMP functions into a QObject that can
be sent over the wire.
Michael Roth [Tue, 19 Jul 2011 19:50:33 +0000 (14:50 -0500)]
qapi: add QMP input visitor
A type of Visiter class that is used to walk a qobject's
structure and assign each entry to the corresponding native C type.
Command marshaling function will use this to pull out QMP command
parameters recieved over the wire and pass them as native arguments
to the corresponding C functions.
Michael Roth [Tue, 19 Jul 2011 19:50:32 +0000 (14:50 -0500)]
qapi: add QAPI visitor core
Base definitions/includes for Visiter interface used by generated
visiter/marshalling code.
Includes a GenericList type. Our lists require an embedded element.
Since these types are generated, if you want to use them in a different
type of data structure, there's no easy way to add another embedded
element. The solution is to have non-embedded lists and that what this is.
Anthony Liguori [Tue, 19 Jul 2011 19:50:29 +0000 (14:50 -0500)]
Add hard build dependency on glib
GLib is an extremely common library that has a portable thread implementation
along with tons of other goodies.
GLib and GObject have a fantastic amount of infrastructure we can leverage in
QEMU including an object oriented programming infrastructure.
Short term, it has a very nice thread pool implementation that we could leverage
in something like virtio-9p. It also has a test harness implementation that
this series will use.
This will cause a build error when files including error.h
don't include qemu-common.h (or compiler.h). Not an issue today
because the only file including it is json-parser.h and it does
include qemu-common.h, but let's get it fixed.
This moves compiler related macros from qemu-common.h to compiler.h.
The reason for this change is that there are simple header files that
depend only on the compiler macros, so including qemu-common.h is overkill.
Besides, qemu-common.h is bloated and will benefit from some splitting.
Please, also note that the QEMU_BUILD_BUG_ON() macro is being fixed to
not use double underscores as a prefix and the license text was added
by Vassili Karpov (malc), who is one of the authors of the new file.
SPARC64: fix VIS1 SIMD signed compare instructions
The destination registers of SIMD signed compare instructions
(fcmp*<16|32>) are not FP registers but general purpose r registers.
Comparisons should be freg_rs1 CMP freg_rs2, that were reversed.
Blue Swirl [Wed, 20 Jul 2011 20:23:01 +0000 (20:23 +0000)]
Merge branch 'for-upstream' of git://git.linaro.org/people/pmaydell/qemu-arm
* 'for-upstream' of git://git.linaro.org/people/pmaydell/qemu-arm:
target-arm: Fix BASEPRI, BASEPRI_MAX, and FAULTMASK access
target-arm: Minimal implementation of performance counters
Revert "Makefile.target: Allow target helpers to be in any *_helper.c file"
Revert "target-arm: Use global env in neon_helper.c helpers"
target-arm: Pass fp status pointer explicitly to neon fp helpers
target-arm: Make VFP binop helpers take pointer to fpstatus, not CPUState
target-arm: Add helper function to generate code to get fpstatus pointer
Revert "target-arm: Use global env in iwmmxt_helper.c helpers"
qxl: upon reset, if spice worker is stopped, the command rings can be not empty
Spice worker does no longer process commands when it is stopped.
Otherwise, it might crash during migration when attempting to process
commands while the guest is not completely loaded.
Kevin Wolf [Tue, 19 Jul 2011 11:01:48 +0000 (13:01 +0200)]
qcow2: Use Qcow2Cache in writeback mode during loadvm/savevm
In snapshotting there is no guest involved, so we can safely use a writeback
mode and do the flushes in the right place (i.e. at the very end). This
improves the time that creating/restoring an internal snapshot takes with an
image in writethrough mode.
qemu-img.c wants to count allocated file size of image. Previously it
counts a single bs->file by 'stat' or Window API. As VMDK introduces
multiple file support, the operation becomes format specific with
platform specific meanwhile.
The functions are moved to block/raw-{posix,win32}.c and qemu-img.c calls
bdrv_get_allocated_file_size to count the bs. And also added VMDK code
to count his own extents.
Add create option 'format', with enums:
monolithicSparse
monolithicFlat
twoGbMaxExtentSparse
twoGbMaxExtentFlat
Each creates a subformat image file. The default is monolithicSparse.
The return type of get_cluster_offset was an offset that use 0 to denote
'not allocated', this will be no longer true for flat extents, as we see
flat extent file as a single huge cluster whose offset is 0 and length
is the whole file length.
So now we use int return value, 0 means success and otherwise offset
invalid.
Cid_update is the flag for updating CID on first write after opening the
image. This should be per image open rather than per program life cycle,
so change it from static var of vmdk_write to a field in BDRVVmdkState.
There are several occurrence of magic number 0x200 as the descriptor
offset within mono sparse image file. This is not the case for images
with separate descriptor file. So a field is added to BDRVVmdkState to
hold the correct value.
Probe as the same behavior as VMware does.
Recognize image as monolithicFlat descriptor file when the file is text
and the first effective line (not '#' leaded comment or space line) is
either 'version=1' or 'version=2'. No space or upper case charactors
accepted.
VMDK: bugfix, align offset to cluster in get_whole_cluster
In get_whole_cluster, the offset is not aligned to cluster when reading
from backing_hd. When the first write to child is not at the cluster
boundary, wrong address data from parent is copied to child.
Hannes Reinecke [Mon, 11 Jul 2011 13:02:24 +0000 (15:02 +0200)]
scsi: Add 'hba_private' to SCSIRequest
'tag' is just an abstraction to identify the command
from the driver. So we should make that explicit by
replacing 'tag' with a driver-defined pointer 'hba_private'.
This saves the lookup for driver handling several commands
in parallel.
'tag' is still being kept for tracing purposes.
Hannes Reinecke [Mon, 11 Jul 2011 13:02:23 +0000 (15:02 +0200)]
iov: Update parameter usage in iov_(to|from)_buf()
iov_to_buf() has an 'offset' parameter, iov_from_buf() hasn't.
This patch adds the missing parameter to iov_from_buf().
It also renames the 'offset' parameter to 'iov_off' to
emphasize it's the offset into the iovec and not the buffer.
Amit Shah [Thu, 7 Jul 2011 12:05:27 +0000 (17:35 +0530)]
virtio-console: Prevent abort()s in case of host chardev close
A host chardev could close just before the guest sends some data to be
written. This will cause an -EPIPE error. This shouldn't be propagated
to virtio-serial-bus.
Ideally we should close the port once -EPIPE is received, but since the
chardev interface doesn't return such meaningful values to its users,
all we get is -1 for any kind of error. Just return 0 for now and wait
for chardevs to return better error messages to act better on the return
messages.
vhost dev stop failed to clear the log field.
Typically not an issue as dev start overwrites this field,
but if logging gets disabled before the following start,
it doesn't so this causes a double free.
Alexander Graf [Sun, 3 Jul 2011 07:44:48 +0000 (09:44 +0200)]
xen_console: fall back to qemu serial device
The new xen_console protocol changed the default xen_console output device
from whatever Qemu chose to whatever xenstore choses and "pty" as fallback.
This is not how Qemu works. It has its own serial redirection semantics. So
it xenstore doesn't contain information on what to do, Qemu is the place to
ask.
xen_console: support the new extended xenstore protocol
Since CS 21994 on xen-unstable.hg and CS 466608f3a32e1f9808acdf832a5843af37e5fcec on qemu-xen-unstable.git, few
changes have been introduced to the PV console xenstore protocol, as
described by the document docs/misc/console.txt under xen-unstable.hg.
From the Qemu point of view, very few modifications are needed to
correctly support the protocol: read from xenstore the "output" node
that tell us what the output of the PV console is going to be.
In case the output is a tty, write to xenstore the device name.
Changes in v2:
- fix error paths: free malloc'ed strings and close the xenstore
connection before returning;
- remove useless snprintf in xenstore_store_pv_console_info if i == 0.
Changes in v3:
- replace xs_daemon_open/xs_daemon_close with xs_open/xs_close.
Changes in v4:
- add a compatibility implementation of xs_open/xs_close.
checkpatch currently errors out, claiming that spaces need to follow
commas. However, this particular style of defining structs is pretty
common in qemu code and very readable. So let's declare it as supported
for the above case.
Register the vkbd backend even when running as device emulator for HVM
guests: it is useful because it doesn't need a frequent timer like usb.
Check whether the XenInput DisplayState has been set in the initialise
state, rather than the input state.
In case the DisplayState hasn't been set and there is no vfb for this
domain, then set the XenInput DisplayState to the default one.
Jan Kiszka [Tue, 21 Jun 2011 20:59:09 +0000 (22:59 +0200)]
xen: Fold CONFIG_XEN_MAPCACHE into CONFIG_XEN
Xen won't be enabled if there is no backend support available for the
host. And that also means the map cache will work. So drop the separate
config switch and move the required stubs over to xen-stub.c.
Jan Kiszka [Tue, 21 Jun 2011 20:59:07 +0000 (22:59 +0200)]
xen: Clean up build system
Introduce CONFIG_XEN_BACKEND so that this new config solely controls the
target-independent backend build and CONFIG_XEN can focus on per-target
building.
Changes since v1:
- take pthread_sigmask() out of the ifdef as it is now common
to both parts.
This fix effectively blocks, in the main thread, the signals handled
by signalfd or the compatibility signal thread.
This way, such signals are received synchronously in the main thread
through sigfd_handler() instead of triggering the signal handler
directly, asynchronously.
Peter Maydell [Wed, 22 Jun 2011 14:40:06 +0000 (15:40 +0100)]
tcg/README: Expand advice on number of TCG ops per target insn
Expand the note on the number of TCG ops generated per target insn,
to be clearer about the range of applicability of the 20 op rule
of thumb. Also add a note about the hard MAX_OP_PER_INSTR limit.
Peter Maydell [Tue, 12 Jul 2011 20:27:15 +0000 (21:27 +0100)]
linux-user/signal.c: Rename s390 target_ucontext fields to fix ia64
The ia64 sys/ucontext.h defines macros 'uc_link', 'uc_sigmask' and
'uc_stack'. Rename the s390 target_ucontext struct members to tuc_*,
bringing them into line with the other targets and fixing a compile
failure on ia64 hosts caused by this clash.
linux-user: make MIPS and ARM eabi use same argument reordering
MIPS uses similar calling convention than ARM eabi, where when using
64-bit values some registers are skipped. This patch makes MIPS and ARM
eabi share the argument reordering code.
This affects ftruncate64, creating insane sized fails (or just failing).
The syscall #123 on SH4 should be "TARGET_NR_cacheflush" instead of
"TARGET_NR_modify_ldt" [1]. The only consequence of this misnaming is
that many "Unsupported syscall" warnings are issued when emulating JIT
compilers.
Stefan Hajnoczi [Sat, 9 Jul 2011 09:22:07 +0000 (10:22 +0100)]
os-posix: set groups properly for -runas
Andrew Griffiths reports that -runas does not set supplementary group
IDs. This means that gid 0 (root) is not dropped when switching to an
unprivileged user.
Add an initgroups(3) call to use the -runas user's /etc/groups
membership to update the supplementary group IDs.
esp: cancel current request only if some request is in flight
This bug was introduced in 94d3f98a3f3caddd7875f9a11776daeb84962a7b:
scsi_cancel_io was checking if some request was pending before trying
to cancel it, while scsi_req_cancel always cancels the request.
This may lead to a crash of Qemu due to dereferencing a NULL pointer,
as exhibited by NetBSD 5.1 installer on MIPS Magnum emulation.
Peter Maydell [Mon, 4 Jul 2011 21:02:46 +0000 (22:02 +0100)]
target-alpha, target-ppc: Remove unnecessary setjmp.h include
Remove the include of setjmp.h from the cpu.h of target-alpha
and target-ppc. This is unnecessary because cpu-defs.h already
includes this header; this change brings these two targets
into line with all the rest.
Stefan Weil [Mon, 4 Jul 2011 18:52:38 +0000 (20:52 +0200)]
Remove unneeded setjmp.h (fix compilation on Debian "lenny")
Some versions of png.h cannot be included after setjmp.h,
even when PNG_SKIP_SETJMP_CHECK was defined.
setjmp.h was included from qemu-common.h and is not needed there.
Removing the include statement fixes compilation of ui/vnc-enc-tight.c
with CONFIG_VNC_PNG defined.
Jan Kiszka [Sat, 2 Jul 2011 07:50:51 +0000 (09:50 +0200)]
tcg: Reload local variables after return from longjmp
Recent compilers look deep into cpu_exec, find longjmp as a noreturn
function and decide to smash some stack variables as they won't be used
again. This may lead to env becoming invalid after return from setjmp,
causing crashes. Fix it by reloading env from cpu_single_env in that
case.
Peter Maydell [Wed, 22 Jun 2011 14:16:32 +0000 (15:16 +0100)]
exec-all.h: Make MAX_OP_PER_INSTR large enough for target-arm's uses
The target-arm frontend's worst-case TCG ops per instr is 194 (and in
general many of the "load multiple registers" ARM instructions generate
more than 100 TCG ops). Raise MAX_OP_PER_INSTR accordingly to avoid
possible buffer overruns.
Since it doesn't make any sense for the "64 bit guest on 32 bit host"
case to have a smaller limit than the normal case, we collapse the
two cases back into each other again.
(This increase costs us about 14K in extra static buffer space and
21K of extra margin at the end of a 32MB codegen buffer.)
Peter Maydell [Wed, 22 Jun 2011 10:58:25 +0000 (11:58 +0100)]
exec.c: Fix calculation of code_gen_buffer_max_size
When calculating the point at which we should not try to put another
TB into the code gen buffer, we have to allow not just for OPC_MAX_SIZE
but OPC_BUF_SIZE. This is because the target translate.c will only
stop when an instruction has put it past the OPC_MAX_SIZE limit, so
we have to include the MAX_OP_PER_INSTR margin which that final insn
might have used.
projects / qemu.git / log