OSDN Git Service
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.
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
Michael Roth [Tue, 19 Jul 2011 19:50:31 +0000 (14:50 -0500)]
qapi: add module init types for qapi
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
Michael Roth [Tue, 19 Jul 2011 19:50:30 +0000 (14:50 -0500)]
qlist: add qlist_first()/qlist_next()
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
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.
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Signed-off-by: Michael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
Luiz Capitulino [Mon, 11 Jul 2011 18:01:57 +0000 (15:01 -0300)]
Error: Fix build when qemu-common.h is not included
Commit
e4ea5e2d0e0e4c5188ab45b66f3195062ae059dc added the use of
the macro GCC_FMT_ATTR to error.h, however compiler.h is not
included by error.h
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.
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
Luiz Capitulino [Mon, 11 Jul 2011 17:24:44 +0000 (14:24 -0300)]
Introduce compiler.h header file
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.
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
Jes Sorensen [Mon, 11 Jul 2011 18:01:09 +0000 (20:01 +0200)]
QMP: add snapshot-blkdev-sync command
Add QMP bits for snapshot_blkdev command. This is the same as
snapshot_blkdev in the human monitor. The command is synchronous.
In the future async commands and or a break down of the functionality
into multiple commands might be added.
Also change the 'snapshot_file' argument to 'snapshot-file' in
the human monitor, so that it matches QMP.
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Luiz Capitulino <lcapitulino@gmail.com>
Blue Swirl [Wed, 20 Jul 2011 22:06:33 +0000 (22:06 +0000)]
Update OpenBIOS images for Sparc32 and Sparc64
Update Sparc32 and Sparc64 OpenBIOS images to SVN revision 1045.
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Blue Swirl [Sun, 3 Jul 2011 08:53:46 +0000 (08:53 +0000)]
Fix unassigned memory access handling
cea5f9a28faa528b6b1b117c9ab2d8828f473fef exposed bugs in unassigned memory
access handling. Fix them by always passing CPUState to the handlers.
Reported-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Blue Swirl [Thu, 14 Jul 2011 15:22:20 +0000 (15:22 +0000)]
Avoid CPU endian memory accesses in devices
Don't compile virtio.c in hwlib, it depends on memory accesses
performed in CPU endianness.
Make loads and stores in CPU endianness unavailable to devices
and poison them to avoid further bugs.
Acked-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Mon, 18 Jul 2011 05:32:18 +0000 (14:32 +0900)]
SPARC64: treat UA2007 ASI_BLK_* as translating ASIs.
UA2007 ASI_BLK_* should be added in is_translating_asi().
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Acked-by: Artyom Tarasenko <atar4qemu@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Blue Swirl [Fri, 15 Jul 2011 20:09:10 +0000 (20:09 +0000)]
checkpatch: Fix bracing false positives on #if
789f88d0b21fedfd4251d56bb7a9fbfbda7a4ac7 only fixed #else,
fix also #if.
Reviewed-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Stefan Weil [Wed, 20 Jul 2011 18:56:36 +0000 (20:56 +0200)]
tcg/mips: Fix regression caused by typo (copy + paste bug)
cppcheck reports an error:
qemu/tcg/mips/tcg-target.c:1487: error: Invalid number of character (()
The unpatched code won't compile on mips hosts starting with commit
cea5f9a28faa528b6b1b117c9ab2d8828f473fef.
Cc: Blue Swirl <blauwirbel@gmail.com>
Signed-off-by: Stefan Weil <weil@mail.berlios.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Hervé Poussineau [Mon, 18 Jul 2011 21:34:22 +0000 (23:34 +0200)]
ds1225y: convert to qdev device, and use it in MIPS Jazz emulation
Signed-off-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Hervé Poussineau [Mon, 18 Jul 2011 21:34:21 +0000 (23:34 +0200)]
ds1225y: use trace framework
Signed-off-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Hervé Poussineau [Mon, 18 Jul 2011 21:34:20 +0000 (23:34 +0200)]
ds1225y: Remove protection stuff, which doesn't belong to this device
Signed-off-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Mon, 18 Jul 2011 06:36:57 +0000 (15:36 +0900)]
SPARC64: add missing break on fmovdcc
"break" is missing on V9 fmovdcc (%icc).
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Mon, 18 Jul 2011 06:00:00 +0000 (15:00 +0900)]
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.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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"
Conflicts:
Makefile.target
Anthony Liguori [Tue, 19 Jul 2011 13:04:35 +0000 (08:04 -0500)]
Merge remote-tracking branch 'agraf/xen-next' into staging
Anthony Liguori [Tue, 19 Jul 2011 13:03:20 +0000 (08:03 -0500)]
Merge remote-tracking branch 'kraxel/usb.19' into staging
Anthony Liguori [Tue, 19 Jul 2011 13:02:35 +0000 (08:02 -0500)]
Merge remote-tracking branch 'spice/spice.v38' into staging
Anthony Liguori [Tue, 19 Jul 2011 12:43:54 +0000 (07:43 -0500)]
Merge remote-tracking branch 'mst/for_anthony' into staging
Anthony Liguori [Tue, 19 Jul 2011 12:43:51 +0000 (07:43 -0500)]
Merge remote-tracking branch 'riku/linux-user-for-upstream' into staging
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.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Amit Shah [Thu, 7 Jul 2011 12:46:13 +0000 (18:16 +0530)]
virtio-serial-bus: Fix trailing \n in error_report string
Markus fixed offenders in the file but one instance sneaked in via
another patch. Fix it.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Amit Shah [Tue, 5 Jul 2011 11:07:49 +0000 (16:37 +0530)]
virtio-console: Add some trace events
Add some trace events for messages passed between the char layer and the
virtio-serial bus.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Amit Shah [Tue, 5 Jul 2011 11:06:39 +0000 (16:36 +0530)]
virtio-serial-bus: Add trace events
Add some trace events for messages passed between the guest and host.
Signed-off-by: Amit Shah <amit.shah@redhat.com>
Michael S. Tsirkin [Mon, 20 Jun 2011 10:42:27 +0000 (13:42 +0300)]
virtio: fix indirect descriptor buffer overflow
We were previously allowing arbitrarily-long indirect descriptors, which
could lead to a buffer overflow in qemu-kvm process.
CVE-2011-2212
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Michael S. Tsirkin [Sun, 26 Jun 2011 13:30:45 +0000 (16:30 +0300)]
xen: move to new pci initializers
move ids to pci info structure
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Anthony PERARD [Wed, 22 Jun 2011 15:58:31 +0000 (16:58 +0100)]
hw/piix_pci.c: Fix PIIX3-xen to initialize ids
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Michael S. Tsirkin [Sun, 26 Jun 2011 13:17:27 +0000 (16:17 +0300)]
pci_ids: tweak names to match linux/pci_ids.h
Sync xen names to ones used by linux. Add
xen platform device id as well.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Michael S. Tsirkin [Tue, 21 Jun 2011 17:34:17 +0000 (20:34 +0300)]
vhost: fix double free on device stop
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.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
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.
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Thu, 30 Jun 2011 17:26:29 +0000 (18:26 +0100)]
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.
Changes in v5:
- fix code style.
[agraf] fix build error due to missing stub
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Alexander Graf [Wed, 29 Jun 2011 06:04:27 +0000 (08:04 +0200)]
checkpatch: don't error out on },{ lines
When having code like this:
static PCIDeviceInfo piix_ide_info[] = {
{
.qdev.name = "piix3-ide",
.qdev.size = sizeof(PCIIDEState),
.qdev.no_user = 1,
.no_hotplug = 1,
.init = pci_piix_ide_initfn,
.vendor_id = PCI_VENDOR_ID_INTEL,
.device_id = PCI_DEVICE_ID_INTEL_82371SB_1,
.class_id = PCI_CLASS_STORAGE_IDE,
},{
.qdev.name = "piix4-ide",
.qdev.size = sizeof(PCIIDEState),
.qdev.no_user = 1,
.no_hotplug = 1,
.init = pci_piix_ide_initfn,
.vendor_id = PCI_VENDOR_ID_INTEL,
.device_id = PCI_DEVICE_ID_INTEL_82371AB,
.class_id = PCI_CLASS_STORAGE_IDE,
},{
/* end of list */
}
};
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.
Reported-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Thu, 30 Jun 2011 14:42:31 +0000 (15:42 +0100)]
xen_disk: treat "aio" as "raw"
Sometimes the toolstack uses "aio" without an additional format
identifier, in such cases use "raw".
Updated in v2:
- fix code style.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Mon, 27 Jun 2011 17:26:06 +0000 (18:26 +0100)]
qemu_ram_ptr_length: take ram_addr_t as arguments
qemu_ram_ptr_length should take ram_addr_t as argument rather than
target_phys_addr_t because is doing comparisons with RAMBlock addresses.
cpu_physical_memory_map should create a ram_addr_t address to pass to
qemu_ram_ptr_length from PhysPageDesc phys_offset.
Remove code after abort() in qemu_ram_ptr_length.
Changes in v2:
- handle 0 size in qemu_ram_ptr_length;
- rename addr1 to raddr;
- initialize raddr to ULONG_MAX.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Mon, 27 Jun 2011 15:10:01 +0000 (16:10 +0100)]
xen_disk: cope with missing xenstore "params" node
When disk is a cdrom and the drive is empty the "params" node in
xenstore might be missing completely: cope with it instead of
segfaulting.
Updated in v2:
- actually removed the strchr(blkdev->params, ':') that caused the
segfault;
- free all the allocated strings from xenstore before returning;
Updated in v3:
- set blkdev fields to NULL after free'ing them.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Fri, 24 Jun 2011 16:36:11 +0000 (17:36 +0100)]
xen: add vkbd support for PV on HVM guests
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.
Changed in v2:
- use qemu_free instead of free;
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Fri, 24 Jun 2011 15:59:46 +0000 (16:59 +0100)]
xen_console: fix memory leak
con_init leaks the string "type", fix it.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Stefano Stabellini [Fri, 24 Jun 2011 14:54:48 +0000 (15:54 +0100)]
xen: enable console and disk backend in HVM mode
Initialize the Xen console backend and the Xen disk backend even when
running in HVM mode so that PV on HVM drivers can connect to them.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
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.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Jan Kiszka [Tue, 21 Jun 2011 20:59:08 +0000 (22:59 +0200)]
xen: Clean up map cache API naming
The map cache is a Xen thing, so its API should make this clear.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
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.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Alexander Graf <agraf@suse.de>
Alexandre Raymond [Wed, 15 Jun 2011 05:20:31 +0000 (01:20 -0400)]
Fix signal handling of SIG_IPI when io-thread is enabled
Both the signal thread (via sigwait()) and the cpu thread (via
a normal signal handler) were attempting to catch SIG_IPI.
This resulted in random freezes under Darwin.
This patch separates SIG_IPI from the rest of the signals handled
by the signal thread, because it is independently caught by the cpu
thread.
Signed-off-by: Alexandre Raymond <cerbere@gmail.com>
Acked-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexandre Raymond [Tue, 14 Jun 2011 14:05:36 +0000 (10:05 -0400)]
Fix signal handling when io-thread is disabled
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.
Signed-off-by: Alexandre Raymond <cerbere@gmail.com>
Acked-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Artyom Tarasenko [Thu, 14 Jul 2011 17:37:06 +0000 (19:37 +0200)]
set ELF_HWCAP for SPARC and SPARC64
setting ELF_HWCAP fixes dynamic library loading for Linux/sparc64
This patch allows loading busybox from Debian 6 initrd
Signed-off-by: Artyom Tarasenko <atar4qemu@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Blue Swirl [Thu, 14 Jul 2011 17:30:43 +0000 (17:30 +0000)]
Sparc: fix FPU and AM enable checks for translation
Translation used incorrectly CPUState fields directly to check
for FPU enable state and 32 bit address masking on Sparc64.
Fix by using TB flags instead.
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:43 +0000 (18:41 +0900)]
SPARC64: C99 comment fix for block-transfer ASIs
Fixed C99 comments on block-tranfer ASIs.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:42 +0000 (18:41 +0900)]
SPARC64: Add JPS1 ASI_BLK_AIU[PS]L ASIs for ldfa and stfa
Support JPS1 little endian block transfer ASIs.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:41 +0000 (18:41 +0900)]
SPARC64: Add UA2007 ASI_BLK_AIU[PS]L? ASIs for stfa
Support UA2007 block store ASIs for stfa instructions.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:40 +0000 (18:41 +0900)]
SPARC64: Add UA2007 ASI_BLK_AIU[PS]L? ASIs for ldfa
Support UA2007 block load ASIs for ldfa instructions.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:39 +0000 (18:41 +0900)]
SPARC64: fp_disabled checks on stfa/stdfa/stqfa
stfa/stdfa/stqfa instructions should raise fp_disabled exceptions
if %pstate.PEF==0 or %fprs.FEF==0.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:38 +0000 (18:41 +0900)]
SPARC64: Implement stfa/stdfa/stqfa instrcutions properly
This patch implements sparcv9 stfa/stdfa/stqfa instructions
with non block-store ASIs.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:37 +0000 (18:41 +0900)]
SPARC64: fp_disabled checks on ldfa/lddfa/ldqfa
ldfa/lddfa/ldqfa instructions should raise fp_disabled exceptions
if %pstate.PEF==0 or %fprs.FEF==0.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Tsuneo Saito [Thu, 14 Jul 2011 09:41:36 +0000 (18:41 +0900)]
SPARC64: Implement ldfa/lddfa/ldqfa instructions properly
This patch implements sparcv9 ldfa/lddfa/ldqfa instructions
with non block-load ASIs.
Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Riku Voipio [Tue, 12 Jul 2011 12:40:51 +0000 (15:40 +0300)]
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).
Cc: Wesley W. Terpstra <terpstra@debian.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Riku Voipio [Tue, 12 Jul 2011 13:01:54 +0000 (16:01 +0300)]
linux-user: correct syscall 123 on sh4
As reported by Cédric VINCENT:
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.
Reported-by: Cédric VINCENT <cedric.vincent@st.com>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
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.
Signed-off-by: Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Acked-by: Chris Wright <chrisw@sous-sol.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Hervé Poussineau [Sat, 9 Jul 2011 14:44:41 +0000 (16:44 +0200)]
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.
Signed-off-by: Hervé Poussineau <hpoussin@reactos.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.
Signed-off-by: Stefan Weil <weil@mail.berlios.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.
Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
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.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:10 +0000 (18:28 +0200)]
spapr: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:09 +0000 (18:28 +0200)]
s390-virtio: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:08 +0000 (18:28 +0200)]
ppc405_uc: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:07 +0000 (18:28 +0200)]
pl080: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:06 +0000 (18:28 +0200)]
msix: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:05 +0000 (18:28 +0200)]
msi: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:04 +0000 (18:28 +0200)]
intel-hda: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Tue, 5 Jul 2011 16:28:03 +0000 (18:28 +0200)]
hpet: use specific endian ld/st_phys
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Alexander Graf [Wed, 6 Jul 2011 07:09:23 +0000 (09:09 +0200)]
exec: add endian specific phys ld/st functions
Device code some times needs to access physical memory and does that
through the ld./st._phys functions. However, these are the exact same
functions that the CPU uses to access memory, which means they will
be endianness swapped depending on the target CPU.
However, devices don't know about the CPU's endianness, but instead
access memory directly using their own interface to the memory bus,
so they need some way to read data with their native endianness.
This patch adds _le and _be functions to ld./st._phys.
Signed-off-by: Alexander Graf <agraf@suse.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Wesley W. Terpstra [Tue, 12 Jul 2011 11:42:00 +0000 (14:42 +0300)]
mips: rlimit codes are not the same
The codes for get/setrlimit differ between linux target platforms.
This patch adds conversion.
This is important else programs (rsyslog, python, ...) can go into a
near infinite loop trying to close all the file descriptors from 0 to
-1.
Signed-off-by: Wesley W. Terpstra <terpstra@debian.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Wesley W. Terpstra [Tue, 12 Jul 2011 11:38:22 +0000 (14:38 +0300)]
mips: rlimit incorrectly converts values
Byte swap was applied in the wrong order with testing for
RLIM_INFINITY. On mips bigendian from an amd64 system this results in
infinity being misinterpretted as 2^31-1.
This is a serious bug because it causes setrlimit stack size to kill
all child processes. This means (for example) that 'make' can run no
children. The mechanism of failure:
1. parent sets stack size rlimit to 'infinity'
2. qemu screws this value up
3. child process fetches stack size as a large (but non-infinite) value
4. qemu tries to allocate stack before execution
5. stack allocation fails (too big) and child process dies
Signed-off-by: Wesley W. Terpstra <terpstra@debian.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Wesley W. Terpstra [Tue, 12 Jul 2011 11:34:23 +0000 (14:34 +0300)]
mips: null pointer deref should segfault
Dereferencing a null pointer causes an exception 0xC (EXCP_AdEL)
instead of EXCP_TLBL. This should also trigger a segfault.
Signed-off-by: Wesley W. Terpstra <terpstra@debian.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Wesley W. Terpstra [Tue, 12 Jul 2011 11:33:23 +0000 (14:33 +0300)]
mips: missing syscall returns wrong errno
Return -TARGET_ENOSYS instead of -ENOSYS from linux-user/main.c
* Caused strange 'Level 2 synchronization messages' instead of
correctly reporting the syscall was missing.
* Made glibc simply fail instead of using older syscalls
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Signed-off-by: Wesley W. Terpstra <terpstra@debian.org>
Wesley W. Terpstra [Tue, 12 Jul 2011 11:32:31 +0000 (14:32 +0300)]
mips: sigaltstack args
The syscall sigaltstack takes two parameters, not zero. This patch
should have no impact as only values above 4 influence the runtime
behaviour. Nevertheless, it is wrong.
Signed-off-by: Wesley W. Terpstra <terpstra@debian.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Peter Maydell [Tue, 28 Jun 2011 11:21:57 +0000 (12:21 +0100)]
linux-user/syscall.c: Enforce pselect6 sigset size restrictions
Enforce the same restriction on the size of the sigset passed to
pselect6 as the Linux kernel does. This is both correct and silences
a gcc 4.6 warning about a write-only variable.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Peter Maydell [Mon, 27 Jun 2011 16:44:52 +0000 (17:44 +0100)]
linux-user: Implement prlimit64 syscall
Implement the prlimit64 syscall.
Slightly modified to apply upstream -Riku
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Peter Maydell [Mon, 27 Jun 2011 16:44:51 +0000 (17:44 +0100)]
linux-user: Add syscall numbers from kernel 2.6.39.2
Add syscall numbers for new syscall numbers; this brings us
into line with Linux 2.6.39.2.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Cédric VINCENT [Wed, 29 Jun 2011 13:09:11 +0000 (15:09 +0200)]
linux-user: Add support for even more FB ioctls
This patch was validated with programs from DirectFB-1.0 and
WebKit/DirectFB.
Signed-off-by: Cédric VINCENT <cedric.vincent@st.com>
Cc: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Cédric VINCENT [Wed, 29 Jun 2011 13:09:10 +0000 (15:09 +0200)]
linux-user: Add support for more VT ioctls
DirectFB-1.0 uses at least two of the four added ioctls, and the two
others were added for completeness. This patch was validated with the
program "vlock -all/-new".
Signed-off-by: Cédric VINCENT <cedric.vincent@st.com>
Cc: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Cédric VINCENT [Wed, 29 Jun 2011 13:09:09 +0000 (15:09 +0200)]
linux-user: Add support for KD...LED ioctls
DirectFB-1.0 uses at least one of the four added ioctls, and the three
others were added for completeness. This patch was validated with the
program "setleds" and the following Makefile:
SETLEDS_INIT = setleds -v -num -caps -scroll
SETLEDS_TESTS = sh -c ' \
setleds -v +num +caps +scroll; \
setleds -v -num -caps -scroll; \
setleds -v +num -caps -scroll; \
setleds -v +num +caps -scroll; \
setleds -v +num +caps +scroll; \
setleds -v -num +caps +scroll; \
setleds -v -num -caps +scroll; \
setleds -v -num -caps -scroll'
SETLEDS_HOST = setleds
SETLEDS_QEMU = "SETLEDS_QEMU not set"
.PHONY: setleds_tests
setleds_tests:
rm -f setleds.host setleds.target
$(SETLEDS_INIT:setleds=$(SETLEDS_HOST))
$(SETLEDS_TESTS:setleds=$(SETLEDS_HOST)) >> setleds.host
$(SETLEDS_INIT:setleds=$(SETLEDS_QEMU))
$(SETLEDS_TESTS:setleds=$(SETLEDS_QEMU)) >> setleds.target
cmp setleds.host setleds.target
Signed-off-by: Cédric VINCENT <cedric.vincent@st.com>
Cc: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Cédric VINCENT [Wed, 29 Jun 2011 10:49:41 +0000 (12:49 +0200)]
arm-semi: Provide access to CLI arguments passed through the "-append" option
This patch basically adapts the new semi-hosting command-line support
-- introduced by Wolfgang Schildbach in the commit
2e8785ac -- for use
in system-mode.
Note that the "arm_cmdline_len" and "host_cmdline_len" variables were
renamed respectively "input_size" and "output_size" because:
* in C, the term "length" is generally used to count the number of
character in a string, not to count the number of bytes in a
buffer (as it is the case here).
* in QEMU, the term "host" is used to name variables that are in
the host address space, not to name variables in the target
address space (as it is the case here).
* in the case of this system-call, the terms "input" and "output"
fit the semantic of the official ARM semi-hosting specification
quite well.
I know renaming can be considered harmful but I do think in this case
the semantic really matters to keep this code more understandable.
Signed-off-by: Cédric VINCENT <cedric.vincent@st.com>
Reviewed-by: Christophe Lyon <christophe.lyon@st.com>
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Paul Brook <paul@codesourcery.com>
Cc: Wolfgang Schildbach <wschi@dolby.com>
Cc: Riku Voipio <riku.voipio@iki.fi>
Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Gerd Hoffmann [Thu, 7 Jul 2011 13:18:50 +0000 (15:18 +0200)]
usb-ohci: raise interrupt on attach
Got lost in commit
618c169b577db64ac6589ad48825d2e11760d1a6,
add it back in. Also fix codestyle while we are at it.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Thu, 7 Jul 2011 13:02:58 +0000 (15:02 +0200)]
usb-hub: remove unused descriptor arrays
Somehow they where left over when converting the hub
to the new usb descriptor infrastructure ...
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Wed, 6 Jul 2011 10:40:28 +0000 (12:40 +0200)]
usb: fixup bluetooth descriptors
Commit
4696425cd05c7baa0a4b469d43ba4b8488bcfc0f changes some
endpoints from isocrounous to interrupt by mistake. Fix it.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Jes Sorensen [Mon, 4 Jul 2011 15:33:05 +0000 (17:33 +0200)]
usb_register_port(): do not set port->opaque and port->index twice
Signed-off-by: Jes Sorensen <Jes.Sorensen@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Tue, 5 Jul 2011 14:58:41 +0000 (16:58 +0200)]
usb: update documentation
Add a paragraph on companion controller mode and a
configuration file which sets it all up for you.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Fri, 1 Jul 2011 09:51:02 +0000 (11:51 +0200)]
ehci: add ich9 controller.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Fri, 1 Jul 2011 07:56:43 +0000 (09:56 +0200)]
ehci: fix port count.
The ICH4 EHCI controller which we emulate has six ports not four.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Fri, 1 Jul 2011 07:48:49 +0000 (09:48 +0200)]
uhci: add ich9 controllers
Add ich9 controllers, Factor out properties to a separate
struct and reference it to reduce duplication.
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Gerd Hoffmann [Fri, 1 Jul 2011 09:45:02 +0000 (11:45 +0200)]
pci: add ich9 usb controller ids
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Hans de Goede [Fri, 24 Jun 2011 18:29:05 +0000 (20:29 +0200)]
usb-ohci: Add support for being a companion controller
To use as a companion controller, use pci-ohci as device and set the
masterbus and num-ports properties, ie:
-device usb-ehci,addr=0b.1,multifunction=on,id=ehci0
-device pci-ohci,addr=0b.0,multifunction=on,masterbus=ehci0.0,num-ports=4
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Hans de Goede [Fri, 24 Jun 2011 15:44:53 +0000 (17:44 +0200)]
usb-uhci: Add support for being a companion controller
To use as a companion controller set the masterbus property.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>