This is the delta between the qemu and qemu-kvm v0.12.5 versions

Signed-off-by: Bruce Rogers <brogers@suse.com>

.gitignore

@@ -48,4 +48,5 @@ pc-bios/bios-pq/status
 pc-bios/vgabios-pq/status
 pc-bios/optionrom/multiboot.bin
 pc-bios/optionrom/multiboot.raw
+pc-bios/optionrom/extboot.bin
 .stgit-*

Changelog

@@ -1,3 +1,260 @@
+version 0.12.5
+ - audio/alsa: Handle SND_PCM_STATE_SETUP in alsa_poll_handler
+ - block: Handle multiwrite errors only when all requests have completed
+ - block: Fix early failure in multiwrite
+ - vpc: Use bdrv_(p)write_sync for metadata writes
+ - vmdk: Use bdrv_(p)write_sync for metadata writes
+ - qcow2: Use bdrv_(p)write_sync for metadata writes
+ - qcow: Use bdrv_(p)write_sync for metadata writes
+ - block: Add bdrv_(p)write_sync
+ - qcow2: Restore L1 entry on l2_allocate failure
+ - block/vdi: Fix image opening and creation for odd disk sizes
+ - block/vpc: Fix conversion from size to disk geometry
+ - qcow2: Remove abort on free_clusters failure
+ - vmdk: Fix COW
+ - qcow2: Fix creation of large images
+ - vmdk: fix double free
+ - qemu-options: add documentation for stdio signal=on|off
+ - target-arm : fix parallel saturated subtraction implementation
+ - target-arm : fix thumb2 parallel add/sub opcode decoding
+ - target-arm: fix addsub/subadd implementation
+ - target-i386: fix xchg rax,r8
+ - block/vvfat.c: fix warnings with _FORTIFY_SOURCE
+ - audio/alsa: Spelling typo (paramters)
+ - target-mips: fix DINSU instruction
+ - Correct definitions for FD_CMD_SAVE and FD_CMD_RESTORE
+ - qcow2: Fix corruption after error in update_refcount
+ - qcow2: Fix corruption after refblock allocation
+ - block: Fix multiwrite with overlapping requests
+ - qcow2: Fix error handling in l2_allocate
+ - qcow2: Clear L2 table cache after write error
+ - ide: Fix ide_dma_cancel
+ - usb-bus: fix no params
+ - Avoid crash on '-usbdevice <device>' without parameters
+ - Fix -usbdevice crash
+ - Fix multiboot compilation
+ - Fix missing symbols in .rel/.rela.plt sections
+ - target-ppc: fix RFI by clearing some bits of MSR
+ - Fix typo in balloon help
+ - arm_timer: fix oneshot mode
+ - arm_timer: reload timer when enabled
+ - qemu-sockets: avoid strlen of NULL pointer
+ - block: fix aio_flush segfaults for read-only protocols (e.g. curl)
+ - virtio-blk: fix barrier support
+ - block: fix sector comparism in multiwrite_req_compare
+ - pci: irq_state vmstate breakage
+ - qemu-img: use the heap instead of the huge stack array for win32
+
+version 0.12.4
+ - Workaround for broken OSS_GETVERSION on FreeBSD, part two (Juergen Lock)
+ - oss: fix fragment setting (malc)
+ - oss: issue OSS_GETVERSION ioctl only when needed (malc)
+ - oss: refactor code around policy setting (malc)
+ - oss: workaround for cases when OSS_GETVERSION is not defined (malc)
+ - block: Free iovec arrays allocated by multiwrite_merge() (Stefan Hajnoczi)
+ - lsi: fix segfault in lsi_command_complete (Gerd Hoffmann)
+ - lsi: pass lsi_request to lsi_reselect (Gerd Hoffmann)
+ - lsi: move dma_len+dma_buf into lsi_request (Gerd Hoffmann)
+ - lsi: move current_dev into lsi_request (Gerd Hoffmann)
+ - lsi: have lsi_request for the whole life time of the request. (Gerd Hoffmann)
+ - lsi: use QTAILQ for lsi_queue (Gerd Hoffmann)
+ - tcp/mips: Change TCG_AREG0 (fp -> s0) (Stefan Weil)
+ - sh_pci: fix memory and I/O access (Aurelien Jarno)
+ - Fix incoming migration with iothread (Marcelo Tosatti)
+ - Fix SIGFPE for vnc display of width/height = 1 (Chris Webb)
+ - net: remove broken net_set_boot_mask() boot device validation (Eduardo Habkost)
+ - qcow2: Remove request from in-flight list after error (Kevin Wolf)
+ - qcow2: Don't ignore immediate read/write failures (Kevin Wolf)
+ - block: Fix multiwrite memory leak in error case (Kevin Wolf)
+ - block: Fix error code in multiwrite for immediate failures (Kevin Wolf)
+ - block: Fix multiwrite error handling (Kevin Wolf)
+ - scsi-disk: fix buffer overflow (Gerd Hoffmann)
+ - qcow2: Rewrite alloc_refcount_block/grow_refcount_table (Kevin Wolf)
+ - qcow2: Factor next_refcount_table_size out (Kevin Wolf)
+ - block: avoid creating too large iovecs in multiwrite_merge (Christoph Hellwig)
+ - json-parser: Fix segfault on malformed input (Kevin Wolf)
+ - linux-user: switch default ppc64 CPU to 970fx from 970 (Aurelien Jarno)
+ - target-sh4: MMU: fix store queue addresses (Aurelien Jarno)
+ - target-sh4: MMU: fix ITLB priviledge check (Aurelien Jarno)
+ - target-sh4: MMU: fix mem_idx computation (Aurelien Jarno)
+ - sh7750: handle MMUCR TI bit (Aurelien Jarno)
+ - UHCI spurious interrut fix (Paul Brook)
+ - tcg/mips: fix branch offset during retranslation (Aurelien Jarno)
+ - tcg/arm: correctly save/restore registers in prologue/epilogue (Aurelien Jarno)
+ - workaround for cmd646 bmdma register access while no dma is active (Igor V. Kovalenko)
+ - Fix corner case in chardev udp: parameter (Jan Kiszka)
+ - Don't set default monitor when there is a mux'ed one (Jan Kiszka)
+ - spelling typo (compatibilty) in hw/fw_cfg.c (Vagrant Cascadian)
+ - fdc: fix drive property handling. (Gerd Hoffmann)
+ - target-i386: fix commit c22549204a6edc431e8e4358e61bd56386ff6957 (TeLeMan)
+ - target-i386: fix SIB decoding with index = 4 (Aurelien Jarno)
+ - Fix segfault with ram_size > 4095M without kvm (Ryan Harper)
+ - target-i386: Fix long jumps/calls in long mode with REX.W set (malc)
+ - target-i386: fix lddqu SSE instruction (Aurelien Jarno)
+ - qemu-char.c: drop debug printfs from qemu_chr_parse_compat (Jan Kiszka)
+ - fix undefined shifts by >32 (Paolo Bonzini)
+ - Fix qemu -net user,hostfwd= example (Aurelien Jarno)
+
+version 0.12.3
+  - kvm: Fix eflags corruption in kvm mode (Jan Kiszka)
+  - qcow2: Fix access after end of array (Kevin Wolf)
+  - ide save/restore pio/atapi cmd transfer fields and io buffer (Marcelo Tosatti)
+  - net: Monitor command set_link finds only VLAN clients, fix (Markus Armbruster)
+  - net: info network shows only VLAN clients, fix (Markus Armbruster)
+  - net: net_check_clients() checks only VLAN clients, fix (Markus Armbruster)
+  - net: Fix bogus "Warning: vlan 0 with no nics" with -device (Markus Armbruster)
+  - net: net_check_clients() runs too early to see -device, fix (Markus Armbruster)
+  - net: Remove unused net_client_uninit() (Markus Armbruster)
+  - don't dereference NULL after failed strdup (Jim Meyering)
+  - virtio-net: fix network stall under load (Tom Lendacky)
+  - json: fix PRId64 on Win32 (Roy Tam)
+  - fix inet_parse typo (Marcelo Tosatti)
+  - iothread: fix vcpu stop with smp tcg (Marcelo Tosatti)
+  - segfault due to buffer overrun in usb-serial (David S. Ahern)
+  - qcow2: Fix signedness bugs (Kevin Wolf)
+  - Do not ignore error, if open file failed (-serial /dev/tty) (Evgeniy Dushistov)
+  - pc-bios: update to newer version of (stable) seabios (Anthony Liguori)
+  - target-mips: fix ROTR and DROTR by zero (Aurelien Jarno)
+  - target-mips: fix CpU exception for coprocessor 0 (Nathan Froyd)
+  - tcg/mips: fix crash in tcg_out_qemu_ld() (Aurelien Jarno)
+  - target-mips: don't call cpu_loop_exit() from helper.c (Aurelien Jarno)
+  - virtio-blk: Fix error cases which ignored rerror/werror (Kevin Wolf)
+  - virtio-blk: Fix restart after read error (Kevin Wolf)
+  - virtio_blk: Factor virtio_blk_handle_request out (Kevin Wolf)
+  - cirrus: Properly re-register cirrus_linear_io_addr on vram unmap (Jan Kiszka)
+  - qcow2: Don't ignore qcow2_alloc_clusters return value (Kevin Wolf)
+  - qcow2: Don't ignore update_refcount return value (Kevin Wolf)
+  - qcow2: Allow updating no refcounts (Kevin Wolf)
+  - qcow2: Improve error handling in update_refcount (Kevin Wolf)
+  - qcow2: Fix error handling in grow_refcount_table (Kevin Wolf)
+  - block: Return original error codes in bdrv_pread/write (Kevin Wolf)
+  - qcow2: Return 0/-errno in qcow2_alloc_cluster_offset (Kevin Wolf)
+  - qcow2: Return 0/-errno in get_cluster_table (Kevin Wolf)
+  - qcow2: Fix error handling in qcow_save_vmstate (Kevin Wolf)
+  - qcow2: Fix error handling in qcow2_grow_l1_table (Kevin Wolf)
+  - win32/sdl: Fix toggle full screen (Herve Poussineau)
+  - win32: pair qemu_memalign() with qemu_vfree() (Herve Poussineau)
+  - vnc_refresh: calling vnc_update_client might free vs (Stefano Stabellini)
+  - Musicpal: Fix descriptor walk in eth_send (Jan Kiszka)
+  - Musicpal: Fix wm8750 I2C address (Jan Kiszka)
+  - fix savevm command without id or tag (Marcelo Tosatti)
+  - reduce number of reinjects on ACK (Gleb Natapov)
+  - QMP: Fix asynchronous events delivery (Luiz Capitulino)
+  - Documentation: Add missing documentation for qdev related command line options (Stefan Weil)
+  - pc: add driver version compat properties (Gerd Hoffmann)
+  - scsi: device version property (Gerd Hoffmann)
+  - ide: device version property (Gerd Hoffmann)
+  - QMP: Emit asynchronous events on all QMP monitors (Adam Litke)
+  - Fix QEMU_WARN_UNUSED_RESULT (Kevin Wolf)
+
+version 0.12.2:
+  - Qemu's internal TFTP server breaks lock-step-iness of TFTP (Milan Plzik)
+  - osdep.c: Fix accept4 fallback (Kevin Wolf)
+  - pc: add rombar to compat properties for pc-0.10 and pc-0.11 (Gerd Hoffmann)
+  - pci: allow loading roms via fw_cfg. (Gerd Hoffmann)
+  - roms: rework rom loading via fw (Gerd Hoffmann)
+  - fw_cfg: rom loader tweaks. (Gerd Hoffmann)
+  - roms: minor fixes and cleanups. (Gerd Hoffmann)
+  - pc: add machine type for 0.12 (Gerd Hoffmann)
+  - loader: more ignores for rom intended to be loaded by the bios (Aurelien Jarno)
+  - vnc_refresh: return if vd->timer is NULL (Stefano Stabellini)
+  - QMP: Don't free async event's 'data' (Luiz Capitulino)
+  - Handle TFTP ERROR from client (Thomas Horsten)
+  - dmg: fix ->open failure (Christoph Hellwig)
+  - virtio-pci: thinko fix (Michael S. Tsirkin)
+  - pc-bios: Update README (SeaBIOS) (Stefan Weil)
+  - vmware_vga: Check cursor dimensions passed from guest to avoid buffer overflow (Roland Dreier)
+  - remove pending exception on vcpu reset. (Gleb Natapov)
+  - Fix CPU topology initialization (Jiri Denemark)
+  - MCE: Fix bug of IA32_MCG_STATUS after system reset (Huang Ying)
+  - linuxboot: fix gdt address calculation (Avi Kivity)
+  - QMP: Drop wrong assert() (Luiz Capitulino)
+  - vnc: Fix artifacts in hextile decoding (Anthony Liguori)
+  - target-i386: Fix "call im" on x86_64 when executing 32-bit code (Aurelien Jarno)
+  - Add missing newline at the end of options list (Michael Tokarev)
+  - Don't load options roms intended to be loaded by the bios in qemu (Avi Kivity)
+  - USB: Improve usbdevice error messages (Scott Tsai)
+  - cpu-all.h: fix cpu_get_real_ticks() #ifdef (Aurelien Jarno)
+  - alpha: fix compile (Blue Swirl)
+  - user_only: compile everything with -fpie (Kirill A. Shutemov)
+  - fdc/sparc32: don't hang on detection under OBP (Artyom Tarasenko)
+  - scsi-disk: Inquiry with allocation length of CDB < 36 (v4) (Artyom Tarasenko)
+  - e1000: fix init values for command register (Michael S. Tsirkin)
+
+version 0.12.1:
+  - loader: fix rom loading at address 0 (fixes target-arm) (Aurelien Jarno)
+  - loader: fix rom_copy (fixes multiboot) (Kevin Wolf)
+
+version 0.12.0:
+
+  - Update to SeaBIOS 0.5.0
+  - e1000: fix device link status in Linux (Anthony Liguori)
+  - monitor: fix QMP for balloon command (Luiz Capitulino)
+  - QMP: Return an empty dict by default (Luiz Capitulino)
+  - QMP: Only handle converted commands (Luiz Capitulino)
+  - pci: support PCI based option rom loading (Gerd Hoffman/Anthony Liguori)
+  - Fix backcompat for hotplug of SCSI controllers (Daniel P. Berrange)
+  - fdc: fix migration from 0.11 (Juan Quintela)
+  - vmware-vga: fix segv on cursor resize. (Dave Airlie)
+  - vmware-vga: various fixes (Dave Airlie/Anthony Liguori)
+  - qdev: improve property error reporting. (Gerd Hoffmann)
+  - fix vga names in default_list (Gerd Hoffmann)
+  - usb-host: check mon before using it. (Gerd Hoffmann)
+  - usb-net: use qdev for -usbdevice (Gerd Hoffmann)
+  - monitor: Catch printing to non-existent monitor (Luiz Capitulino)
+  - Avoid permanently disabled QEMU monitor when UNIX migration fails (Daniel P. Berrange)
+  - Fix loading of ELF multiboot kernels (Kevin Wolf)
+  - qemu-io: Fix memory leak (Kevin Wolf)
+  - Fix thinko in linuxboot.S (Paolo Bonzini)
+  - target-i386: Fix evaluation of DR7 register (Jan Kiszka)
+  - vnc: hextile: do not generate ForegroundSpecified and SubrectsColoured tiles (Anthony Liguori)
+  - S390: Bail out without KVM (Alexander Graf)
+  - S390: Don't tell guest we're updating config space (Alexander Graf)
+  - target-s390: Fail on unknown instructions (Alexander Graf)
+  - osdep: Fix runtime failure on older Linux kernels (Andre Przywara)
+  - Fix a make -j race (Juergen Lock)
+  - target-alpha: Fix generic ctz64. (Richard Henderson)
+  - s390: Fix buggy assignment (Stefan Weil)
+  - target-mips: fix user-mode emulation startup (Nathan Froyd)
+  - target-i386: Update CPUID feature set for TCG (Andre Przywara)
+  - s390: fix build on 32 bit host (Michael S. Tsirkin)
+	
+version 0.12.0-rc2:
+
+  - v2: properly save kvm system time msr registers (Glauber Costa)
+  - convert more monitor commands to qmp (Luiz Capitulino)
+  - vnc: fix capslock tracking logic. (Gerd Hoffmann)
+  - QemuOpts: allow larger option values. (Gerd Hoffmann)
+  - scsi: fix drive hotplug. (Gerd Hoffmann)
+  - pci: don't hw_error() when no slot is available. (Gerd Hoffmann)
+  - pci: don't abort() when trying to hotplug with acpi off. (Gerd Hoffmann)
+  - allow default devices to be implemented in config file (Gerd Hoffman)
+  - vc: colorize chardev title line with blue background. (Gerd Hoffmann)
+  - chardev: make chardevs specified in config file work. (Gerd Hoffmann)
+  - qdev: also match bus name for global properties (Gerd Hoffmann)
+  - qdev: add command line option to set global defaults for properties. (Gerd Hoffmann)
+  - kvm: x86: Save/restore exception_index (Jan Kiszka)
+  - qdev: Replace device names containing whitespace (Markus Armbruster)
+  - fix rtc-td-hack on host without high-res timers (Gleb Natapov)
+  - virtio: verify features on load (Michael S. Tsirkin)
+  - vmware_vga: add rom file so that it boots. (Dave Airlie)
+  - Do not abort on qemu_malloc(0) in production builds (Anthony Liguori)
+  - Fix ARM userspace strex implementation. (Paul Brook)
+  - qemu: delete rule target on error (Michael S. Tsirkin)
+  - QMP: add human-readable description to error response (Markus Armbruster)
+  - convert more monitor commands to QError (Markus Armbruster)
+  - monitor: Fix double-prompt after "change vnc passwd BLA" (Markus Armbruster)
+  - monitor: do_cont(): Don't ask for passwords (Luiz Capitulino)
+  - monitor: Introduce 'block_passwd' command (Luiz Capitulino)
+  - pci: interrupt disable bit support (Michael S. Tsirkin)
+  - pci: interrupt status bit implementation (Michael S. Tsirkin)
+  - pci: prepare irq code for interrupt state (Michael S. Tsirkin)
+  - msix: function mask support (Michael S. Tsirkin)
+  - msix: macro rename for function mask support (Michael S. Tsirkin)
+  - cpuid: Fix multicore setup on Intel (Andre Przywara)
+  - kvm: x86: Fix initial kvm_has_msr_star (Jan Kiszka)
+  - Update OpenBIOS images to r640 (Aurelien Jarno)	
+
 version 0.10.2:
 
   - fix savevm/loadvm (Anthony Liguori)

EXTERNAL_DEPENDENCIES

@@ -0,0 +1,2 @@
+seabios 9fb3f4d950744e97cc655b7d7b523d8bf101e4a0
+vgabios 6e62666cfc19e7fd45dd0d7c3ad62fd8d0b5f67a

KVM_VERSION

@@ -0,0 +1 @@
+qemu-kvm-0.12.5

Makefile

@@ -63,6 +63,18 @@ config-host.h-timestamp: config-host.mak
 
 SUBDIR_RULES=$(patsubst %,subdir-%, $(TARGET_DIRS))
 
+ifeq ($(KVM_KMOD),yes)
+
+.PHONEY: kvm-kmod
+
+all: kvm-kmod
+
+kvm-kmod:
+	$(call quiet-command,$(MAKE) $(SUBDIR_MAKEFLAGS) -C kvm/kernel V="$(V)" )
+
+
+endif
+
 subdir-%: $(GENERATED_HEADERS)
 	$(call quiet-command,$(MAKE) $(SUBDIR_MAKEFLAGS) -C $* V="$(V)" TARGET_DIR="$*/" all,)
 
@@ -70,7 +82,7 @@ $(filter %-softmmu,$(SUBDIR_RULES)): libqemu_common.a
 
 $(filter %-user,$(SUBDIR_RULES)): libuser.a
 
-libuser.a:
+libuser.a: $(GENERATED_HEADERS)
 	$(call quiet-command,$(MAKE) $(SUBDIR_MAKEFLAGS) -C libuser V="$(V)" TARGET_DIR="libuser/" all,)
 
 ROMSUBDIR_RULES=$(patsubst %,romsubdir-%, $(ROMS))
@@ -81,6 +93,12 @@ ALL_SUBDIRS=$(TARGET_DIRS) $(patsubst %,pc-bios/%, $(ROMS))
 
 recurse-all: $(SUBDIR_RULES) $(ROMSUBDIR_RULES)
 
+#######################################################################
+# QObject
+qobject-obj-y = qint.o qstring.o qdict.o qlist.o qfloat.o qbool.o
+qobject-obj-y += qjson.o json-lexer.o json-streamer.o json-parser.o
+qobject-obj-y += qerror.o
+
 #######################################################################
 # block-obj-y is code used by both qemu system emulation and qemu-img
 
@@ -88,6 +106,7 @@ block-obj-y = cutils.o cache-utils.o qemu-malloc.o qemu-option.o module.o
 block-obj-y += nbd.o block.o aio.o aes.o osdep.o
 block-obj-$(CONFIG_POSIX) += posix-aio-compat.o
 block-obj-$(CONFIG_LINUX_AIO) += linux-aio.o
+block-obj-$(CONFIG_POSIX) += compatfd.o
 
 block-nested-y += cow.o qcow.o vdi.o vmdk.o cloop.o dmg.o bochs.o vpc.o vvfat.o
 block-nested-y += qcow2.o qcow2-refcount.o qcow2-cluster.o qcow2-snapshot.o
@@ -120,6 +139,7 @@ net-obj-y += $(addprefix net/, $(net-nested-y))
 
 obj-y = $(block-obj-y)
 obj-y += $(net-obj-y)
+obj-y += $(qobject-obj-y)
 obj-y += readline.o console.o
 
 obj-y += tcg-runtime.o host-utils.o
@@ -152,8 +172,6 @@ obj-y += buffered_file.o migration.o migration-tcp.o qemu-sockets.o
 obj-y += qemu-char.o aio.o savevm.o
 obj-y += msmouse.o ps2.o
 obj-y += qdev.o qdev-properties.o
-obj-y += qint.o qstring.o qdict.o qlist.o qfloat.o qbool.o json-lexer.o
-obj-y += json-streamer.o json-parser.o qjson.o qerror.o
 obj-y += qemu-config.o block-migration.o
 
 obj-$(CONFIG_BRLAPI) += baum.o
@@ -230,18 +248,18 @@ libqemu_common.a: $(obj-y)
 
 qemu-img.o: qemu-img-cmds.h
 
-qemu-img$(EXESUF): qemu-img.o qemu-tool.o $(block-obj-y)
+qemu-img$(EXESUF): qemu-img.o qemu-tool.o $(block-obj-y) $(qobject-obj-y)
 
-qemu-nbd$(EXESUF):  qemu-nbd.o qemu-tool.o $(block-obj-y)
+qemu-nbd$(EXESUF):  qemu-nbd.o qemu-tool.o $(block-obj-y) $(qobject-obj-y)
 
-qemu-io$(EXESUF):  qemu-io.o qemu-tool.o cmd.o $(block-obj-y)
+qemu-io$(EXESUF):  qemu-io.o qemu-tool.o cmd.o $(block-obj-y) $(qobject-obj-y)
 
 qemu-img-cmds.h: $(SRC_PATH)/qemu-img-cmds.hx
 	$(call quiet-command,sh $(SRC_PATH)/hxtool -h < $< > $@,"  GEN   $@")
 
 check-qint: check-qint.o qint.o qemu-malloc.o
 check-qstring: check-qstring.o qstring.o qemu-malloc.o
-check-qdict: check-qdict.o qdict.o qint.o qstring.o qemu-malloc.o
+check-qdict: check-qdict.o qdict.o qint.o qstring.o qbool.o qemu-malloc.o qlist.o
 check-qlist: check-qlist.o qlist.o qint.o qemu-malloc.o
 check-qfloat: check-qfloat.o qfloat.o qemu-malloc.o
 check-qjson: check-qjson.o qfloat.o qint.o qdict.o qstring.o qlist.o qbool.o qjson.o json-streamer.o json-lexer.o json-parser.o qemu-malloc.o
@@ -278,6 +296,8 @@ pxe-ne2k_pci.bin pxe-pcnet.bin \
 pxe-rtl8139.bin pxe-virtio.bin \
 bamboo.dtb petalogix-s3adsp1800.dtb \
 multiboot.bin linuxboot.bin
+BLOBS += extboot.bin
+BLOBS += vapic.bin
 else
 BLOBS=
 endif
@@ -300,7 +320,12 @@ endif
 ifneq ($(BLOBS),)
 	$(INSTALL_DIR) "$(DESTDIR)$(datadir)"
 	set -e; for x in $(BLOBS); do \
+	    if [ -f $(SRC_PATH)/pc-bios/$$x ];then \
 		$(INSTALL_DATA) $(SRC_PATH)/pc-bios/$$x "$(DESTDIR)$(datadir)"; \
+	    fi \
+	    ; if [ -f pc-bios/optionrom/$$x ];then \
+		$(INSTALL_DATA) pc-bios/optionrom/$$x "$(DESTDIR)$(datadir)"; \
+	    fi \
 	done
 endif
 	$(INSTALL_DIR) "$(DESTDIR)$(datadir)/keymaps"
@@ -310,6 +335,9 @@ endif
 	for d in $(TARGET_DIRS); do \
 	$(MAKE) -C $$d $@ || exit 1 ; \
         done
+ifeq ($(KVM_KMOD),yes)
+	$(MAKE) -C kvm/kernel $@
+endif
 
 # various test targets
 test speed: all
@@ -430,6 +458,7 @@ tarbin:
 	$(datadir)/pxe-rtl8139.bin \
 	$(datadir)/pxe-pcnet.bin \
 	$(datadir)/pxe-e1000.bin \
+	$(datadir)/extboot.bin \
 	$(docdir)/qemu-doc.html \
 	$(docdir)/qemu-tech.html \
 	$(mandir)/man1/qemu.1 \

Makefile.hw

@@ -25,7 +25,9 @@ obj-$(CONFIG_ESCC) += escc.o
 # PCI watchdog devices
 obj-y += wdt_i6300esb.o
 
-obj-y += msix.o
+# MSI-X depends on kvm for interrupt injection,
+# so moved it from Makefile.hw to Makefile.target for now
+# obj-y += msix.o
 
 # PCI network cards
 obj-y += ne2000.o

Makefile.target

@@ -30,6 +30,8 @@ LIBS+=-lm
 
 kvm.o kvm-all.o: QEMU_CFLAGS+=$(KVM_CFLAGS)
 
+CFLAGS += $(KVM_CFLAGS)
+
 config-target.h: config-target.h-timestamp
 config-target.h-timestamp: config-target.mak
 
@@ -40,12 +42,18 @@ all: $(PROGS)
 
 #########################################################
 # cpu emulator library
-libobj-y = exec.o translate-all.o cpu-exec.o translate.o
+libobj-y = exec.o cpu-exec.o
-libobj-y += tcg/tcg.o
+libobj-$(CONFIG_NO_CPU_EMULATION) += fake-exec.o
+libobj-$(CONFIG_CPU_EMULATION) += translate-all.o translate.o
+libobj-$(CONFIG_CPU_EMULATION) += tcg/tcg.o
 libobj-$(CONFIG_SOFTFLOAT) += fpu/softfloat.o
 libobj-$(CONFIG_NOSOFTFLOAT) += fpu/softfloat-native.o
 libobj-y += op_helper.o helper.o
 libobj-$(CONFIG_NEED_MMU) += mmu.o
+
+libobj-$(CONFIG_KVM) += kvm-tpr-opt.o
+libobj-$(CONFIG_KVM) += qemu-kvm-helper.o
+
 libobj-$(TARGET_ARM) += neon_helper.o iwmmxt_helper.o
 libobj-$(TARGET_ALPHA) += alpha_palcode.o
 
@@ -82,6 +90,8 @@ op_helper.o cpu-exec.o: QEMU_CFLAGS += $(HELPER_CFLAGS)
 # cpu_signal_handler() in cpu-exec.c.
 signal.o: QEMU_CFLAGS += $(HELPER_CFLAGS)
 
+qemu-kvm-helper.o: QEMU_CFLAGS += $(HELPER_CFLAGS)
+
 #########################################################
 # Linux user emulator target
 
@@ -158,6 +168,10 @@ obj-y = vl.o async.o monitor.o pci.o pci_host.o pcie_host.o machine.o gdbstub.o
 # need to fix this properly
 obj-y += virtio-blk.o virtio-balloon.o virtio-net.o virtio-console.o virtio-pci.o
 obj-$(CONFIG_KVM) += kvm.o kvm-all.o
+# MSI-X depends on kvm for interrupt injection,
+# so moved it from Makefile.hw to Makefile.target for now
+obj-y += msix.o
+
 obj-$(CONFIG_ISA_MMIO) += isa_mmio.o
 LIBS+=-lz
 
@@ -194,12 +208,25 @@ obj-i386-y += fdc.o mc146818rtc.o serial.o i8259.o i8254.o pcspk.o pc.o
 obj-i386-y += cirrus_vga.o apic.o ioapic.o parallel.o acpi.o piix_pci.o
 obj-i386-y += usb-uhci.o vmmouse.o vmport.o vmware_vga.o hpet.o
 obj-i386-y += device-hotplug.o pci-hotplug.o smbios.o wdt_ib700.o
+obj-i386-y += extboot.o
 obj-i386-y += ne2000-isa.o
+obj-i386-y += testdev.o
+
+obj-i386-$(CONFIG_KVM_PIT) += i8254-kvm.o
+obj-i386-$(CONFIG_KVM_DEVICE_ASSIGNMENT) += device-assignment.o
+
+# Hardware support
+obj-ia64-y += ide.o pckbd.o vga.o $(SOUND_HW) dma.o $(AUDIODRV)
+obj-ia64-y += fdc.o mc146818rtc.o serial.o i8259.o ipf.o
+obj-ia64-y += cirrus_vga.o parallel.o acpi.o piix_pci.o
+obj-ia64-y += usb-uhci.o
+obj-ia64-$(CONFIG_KVM_DEVICE_ASSIGNMENT) += device-assignment.o
 
 # shared objects
 obj-ppc-y = ppc.o ide/core.o ide/qdev.o ide/isa.o ide/pci.o ide/macio.o
 obj-ppc-y += ide/cmd646.o
 obj-ppc-y += vga.o vga-pci.o $(sound-obj-y) dma.o openpic.o
+obj-ppc-y += cirrus_vga.o
 # PREP target
 obj-ppc-y += pckbd.o serial.o i8259.o i8254.o fdc.o mc146818rtc.o
 obj-ppc-y += prep_pci.o ppc_prep.o ne2000-isa.o
@@ -295,6 +322,11 @@ obj-m68k-y += m68k-semi.o dummy_m68k.o
 
 obj-s390x-y = s390-virtio-bus.o s390-virtio.o
 
+ifeq ($(TARGET_ARCH), ia64)
+firmware.o: firmware.c
+	$(CC) $(HELPER_CFLAGS) $(CPPFLAGS) $(BASE_CFLAGS) -c -o $@ $<
+endif
+
 main.o vl.o: QEMU_CFLAGS+=$(GPROF_CFLAGS)
 
 vl.o: QEMU_CFLAGS+=$(SDL_CFLAGS)

Makefile.user

@@ -2,10 +2,15 @@
 
 include ../config-host.mak
 include $(SRC_PATH)/rules.mak
+-include config.mak
 
 .PHONY: all
 
-VPATH=$(SRC_PATH)
+# Do not take %.o from $(SRC_PATH), only %.c and %.h
+# All %.o for user targets should be built with -fpie, when
+# configured with --enable-user-pie, so we don't want to
+# take %.o from $(SRC_PATH), since they built without -fpie
+vpath %.c %.h $(SRC_PATH)
 
 QEMU_CFLAGS+=-I..
 

QMP/README

@@ -4,45 +4,57 @@
 Introduction
 -------------
 
-The QEMU Monitor Protocol (QMP) is a JSON[1] based protocol for QEMU.
+The QEMU Monitor Protocol (QMP) allows applications to communicate with
+QEMU's Monitor.
 
-By using it applications can control QEMU in reliable and "parseable" way,
+QMP is JSON[1] based and has the following features:
-QMP also provides asynchronous events support.
+
+- Lightweight, text-based, easy to parse data format
+- Asynchronous events support 
+- Stability
 
 For more information, please, refer to the following files:
 
-o qmp-spec.txt    QEMU Monitor Protocol current draft specification
+o qmp-spec.txt    QEMU Monitor Protocol current specification
 o qmp-events.txt  List of available asynchronous events
 
 There are also two simple Python scripts available:
 
 o qmp-shell       A shell
-o vm-info         Show some informations about the Virtal Machine
+o vm-info         Show some information about the Virtual Machine
 
 [1] http://www.json.org
 
 Usage
 -----
 
-To enable QMP, QEMU has to be started in "control mode". This is done
+To enable QMP, QEMU has to be started in "control mode". There are
-by passing the flag "control" to the "-monitor" command-line option.
+two ways of doing this, the simplest one is using the the '-qmp'
+command-line option.
 
 For example:
 
-$ qemu [...] -monitor control,tcp:localhost:4444,server
+$ qemu [...] -qmp tcp:localhost:4444,server
 
 Will start QEMU in control mode, waiting for a client TCP connection
 on localhost port 4444.
 
-To manually test it you can connect with telnet and issue commands:
+It is also possible to use the '-mon' command-line option to have
+more complex combinations. Please, refer to the QEMU's manpage for
+more information.
+
+Simple Testing
+--------------
+
+To manually test QMP one can connect with telnet and issue commands:
 
 $ telnet localhost 4444
-Trying ::1...
+Trying 127.0.0.1...
 Connected to localhost.
 Escape character is '^]'.
 {"QMP": {"capabilities": []}}
 { "execute": "query-version" }
-{"return": "0.11.50"}
+{"return": {"qemu": "0.11.50", "package": ""}}
 
 Contact
 -------

QMP/qmp-spec.txt

@@ -1,4 +1,4 @@
-           QEMU Monitor Protocol Draft Specification - Version 0.1
+           QEMU Monitor Protocol Specification - Version 0.1
 
 1. Introduction
 ===============
@@ -27,9 +27,9 @@ the JSON standard:
 
 http://www.ietf.org/rfc/rfc4627.txt
 
-For convenience, json-objects mentioned in this document will have its members
+For convenience, json-object members and json-array elements mentioned in
-in a certain order. However, in real protocol usage json-objects members can
+this document will be in a certain order. However, in real protocol usage
-be in ANY order, thus no particular order should be assumed.
+they can be in ANY order, thus no particular order should be assumed.
 
 2.1 General Definitions
 -----------------------
@@ -85,12 +85,13 @@ without errors.
 
 The format is:
 
-{ "return": json-value, "id": json-value }
+{ "return": json-object, "id": json-value }
 
  Where,
 
 - The "return" member contains the command returned data, which is defined
-  in a per-command basis or "OK" if the command does not return data
+  in a per-command basis or an empty json-object if the command does not
+  return data
 - The "id" member contains the transaction identification associated
   with the command execution (if issued by the Client)
 
@@ -102,13 +103,16 @@ completed because of an error condition.
 
 The format is:
 
-{ "error": { "class": json-string, "data": json-value }, "id": json-value }
+{ "error": { "class": json-string, "data": json-object, "desc": json-string },
+  "id": json-value }
 
  Where,
 
 - The "class" member contains the error class name (eg. "ServiceUnavailable")
 - The "data" member contains specific error data and is defined in a
   per-command basis, it will be an empty json-object if the error has no data
+- The "desc" member is a human-readable error message. Clients should
+  not attempt to parse this message.
 - The "id" member contains the transaction identification associated with
   the command execution (if issued by the Client)
 
@@ -124,7 +128,7 @@ to the Client at any time. They are called 'asynchronous events'.
 
 The format is:
 
-{ "event": json-string, "data": json-value,
+{ "event": json-string, "data": json-object,
   "timestamp": { "seconds": json-number, "microseconds": json-number } }
 
  Where,
@@ -132,7 +136,7 @@ The format is:
 - The "event" member contains the event's name
 - The "data" member contains event specific data, which is defined in a
   per-event basis, it is optional
-- The "timestamp" member contains the exact time of when the event ocurred
+- The "timestamp" member contains the exact time of when the event occurred
   in the Server. It is a fixed json-object with time in seconds and
   microseconds
 
@@ -154,19 +158,20 @@ S: {"QMP": {"capabilities": []}}
 ---------------------------
 
 C: { "execute": "stop" }
-S: {"return": "OK"}
+S: {"return": {}}
 
 3.3 KVM information
 -------------------
 
-C: {"execute": "query-kvm", "id": "example"}
+C: { "execute": "query-kvm", "id": "example" }
-S: {"return": "enabled", "id": "example"}
+S: {"return": {"enabled": true, "present": true}, "id": "example"}
 
 3.4 Parsing error
 ------------------
 
 C: { "execute": }
-S: {"error": {"class": "JSONParsing", "data": {}}}
+S: {"error": {"class": "JSONParsing", "desc": "Invalid JSON syntax", "data":
+{}}}
 
 3.5 Powerdown event
 -------------------
@@ -174,19 +179,25 @@ S: {"error": {"class": "JSONParsing", "data": {}}}
 S: {"timestamp": {"seconds": 1258551470, "microseconds": 802384}, "event":
 "POWERDOWN"}
 
-4. Notes to Client implementors
+4. Compatibility Considerations
--------------------------------
+--------------------------------
 
-4.1 It is recommended to always start the Server in pause mode, thus the
+In order to achieve maximum compatibility between versions, Clients must not 
-    Client is able to perform any setup procedure without the risk of
+assume any particular:
-    race conditions and related problems
 
-4.2 It is recommended to always check the capabilities json-array, issued
+- Size of json-objects or length of json-arrays
-    with the greeting message, at connection time
+- Order of json-object members or json-array elements
+- Amount of errors generated by a command, that is, new errors can be added
+  to any existing command in newer versions of the Server
 
-4.3 Json-objects or json-arrays mentioned in this document are not fixed
+Additionally, Clients should always:
-    and no particular size or number of members/elements should be assumed.
-    New members/elements can be added at any time.
 
-4.4 No particular order of json-objects members should be assumed, they
+- Check the capabilities json-array at connection time
-    can change at any time
+- Check the availability of commands with 'query-commands' before issuing them
+
+5. Recommendations to Client implementors
+-----------------------------------------
+
+5.1 The Server should be always started in pause mode, thus the Client is
+    able to perform any setup procedure without the risk of race conditions
+    and related problems

VERSION

@@ -1 +1 @@
-0.11.50
+0.12.5

aio.c

@@ -113,7 +113,9 @@ void qemu_aio_flush(void)
         qemu_aio_wait();
 
         QLIST_FOREACH(node, &aio_handlers, node) {
-            ret |= node->io_flush(node->opaque);
+            if (node->io_flush) {
+                ret |= node->io_flush(node->opaque);
+            }
         }
     } while (qemu_bh_poll() || ret > 0);
 }

audio/alsaaudio.c

@@ -213,6 +213,10 @@ static void alsa_poll_handler (void *opaque)
 
     state = snd_pcm_state (hlp->handle);
     switch (state) {
+    case SND_PCM_STATE_SETUP:
+        alsa_recover (hlp->handle);
+        break;
+
     case SND_PCM_STATE_XRUN:
         alsa_recover (hlp->handle);
         break;
@@ -665,7 +669,7 @@ static int alsa_open (int in, struct alsa_params_req *req,
         (obt->fmt != req->fmt ||
          obt->nchannels != req->nchannels ||
          obt->freq != req->freq)) {
-        dolog ("Audio paramters for %s\n", typ);
+        dolog ("Audio parameters for %s\n", typ);
         alsa_dump_info (req, obt);
     }
 

audio/ossaudio.c

@@ -38,6 +38,10 @@
 #define AUDIO_CAP "oss"
 #include "audio_int.h"
 
+#if defined OSS_GETVERSION && defined SNDCTL_DSP_POLICY
+#define USE_DSP_POLICY
+#endif
+
 typedef struct OSSVoiceOut {
     HWVoiceOut hw;
     void *pcm_buf;
@@ -236,14 +240,39 @@ static void oss_dump_info (struct oss_params *req, struct oss_params *obt)
 }
 #endif
 
+#ifdef USE_DSP_POLICY
+static int oss_get_version (int fd, int *version, const char *typ)
+{
+    if (ioctl (fd, OSS_GETVERSION, &version)) {
+#if defined(__FreeBSD__) || defined(__FreeBSD_kernel__)
+        /*
+         * Looks like atm (20100109) FreeBSD knows OSS_GETVERSION
+         * since 7.x, but currently only on the mixer device (or in
+         * the Linuxolator), and in the native version that part of
+         * the code is in fact never reached so the ioctl fails anyway.
+         * Until this is fixed, just check the errno and if its what
+         * FreeBSD's sound drivers return atm assume they are new enough.
+         */
+        if (errno == EINVAL) {
+            *version = 0x040000;
+            return 0;
+        }
+#endif
+        oss_logerr2 (errno, typ, "Failed to get OSS version\n");
+        return -1;
+    }
+    return 0;
+}
+#endif
+
 static int oss_open (int in, struct oss_params *req,
                      struct oss_params *obt, int *pfd)
 {
     int fd;
-    int version;
     int oflags = conf.exclusive ? O_EXCL : 0;
     audio_buf_info abinfo;
     int fmt, freq, nchannels;
+    int setfragment = 1;
     const char *dspname = in ? conf.devpath_in : conf.devpath_out;
     const char *typ = in ? "ADC" : "DAC";
 
@@ -281,27 +310,30 @@ static int oss_open (int in, struct oss_params *req,
         goto err;
     }
 
-    if (ioctl (fd, OSS_GETVERSION, &version)) {
+#ifdef USE_DSP_POLICY
-        oss_logerr2 (errno, typ, "Failed to get OSS version\n");
+    if (conf.policy >= 0) {
-        version = 0;
+        int version;
-    }
 
-    if (conf.debug) {
+        if (!oss_get_version (fd, &version, typ)) {
-        dolog ("OSS version = %#x\n", version);
+            if (conf.debug) {
-    }
+                dolog ("OSS version = %#x\n", version);
+            }
 
-#ifdef SNDCTL_DSP_POLICY
+            if (version >= 0x040000) {
-    if (conf.policy >= 0 && version >= 0x040000) {
+                int policy = conf.policy;
-        int policy = conf.policy;
+                if (ioctl (fd, SNDCTL_DSP_POLICY, &policy)) {
-        if (ioctl (fd, SNDCTL_DSP_POLICY, &policy)) {
+                    oss_logerr2 (errno, typ,
-            oss_logerr2 (errno, typ, "Failed to set timing policy to %d\n",
+                                 "Failed to set timing policy to %d\n",
-                         conf.policy);
+                                 conf.policy);
-            goto err;
+                    goto err;
+                }
+                setfragment = 0;
+            }
         }
     }
-    else
 #endif
-    {
+
+    if (setfragment) {
         int mmmmssss = (req->nfrags << 16) | ctz32 (req->fragsize);
         if (ioctl (fd, SNDCTL_DSP_SETFRAGMENT, &mmmmssss)) {
             oss_logerr2 (errno, typ, "Failed to set buffer length (%d, %d)\n",
@@ -857,7 +889,7 @@ static struct audio_option oss_options[] = {
         .valp  = &conf.exclusive,
         .descr = "Open device in exclusive mode (vmix wont work)"
     },
-#ifdef SNDCTL_DSP_POLICY
+#ifdef USE_DSP_POLICY
     {
         .name  = "POLICY",
         .tag   = AUD_OPT_INT,

block.c

@@ -26,6 +26,7 @@
 #include "monitor.h"
 #include "block_int.h"
 #include "module.h"
+#include "qemu-objects.h"
 
 #ifdef CONFIG_BSD
 #include <sys/types.h>
@@ -451,6 +452,8 @@ int bdrv_open2(BlockDriverState *bs, const char *filename, int flags,
             (flags & (BDRV_O_CACHE_MASK|BDRV_O_NATIVE_AIO));
     else
         open_flags = flags & ~(BDRV_O_FILE | BDRV_O_SNAPSHOT);
+
+    bs->open_flags = open_flags;
     if (use_bdrv_whitelist && !bdrv_is_whitelisted(drv))
         ret = -ENOTSUP;
     else
@@ -690,6 +693,7 @@ int bdrv_pread(BlockDriverState *bs, int64_t offset,
     uint8_t tmp_buf[BDRV_SECTOR_SIZE];
     int len, nb_sectors, count;
     int64_t sector_num;
+    int ret;
 
     count = count1;
     /* first read to align to sector start */
@@ -698,8 +702,8 @@ int bdrv_pread(BlockDriverState *bs, int64_t offset,
         len = count;
     sector_num = offset >> BDRV_SECTOR_BITS;
     if (len > 0) {
-        if (bdrv_read(bs, sector_num, tmp_buf, 1) < 0)
+        if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
-            return -EIO;
+            return ret;
         memcpy(buf, tmp_buf + (offset & (BDRV_SECTOR_SIZE - 1)), len);
         count -= len;
         if (count == 0)
@@ -711,8 +715,8 @@ int bdrv_pread(BlockDriverState *bs, int64_t offset,
     /* read the sectors "in place" */
     nb_sectors = count >> BDRV_SECTOR_BITS;
     if (nb_sectors > 0) {
-        if (bdrv_read(bs, sector_num, buf, nb_sectors) < 0)
+        if ((ret = bdrv_read(bs, sector_num, buf, nb_sectors)) < 0)
-            return -EIO;
+            return ret;
         sector_num += nb_sectors;
         len = nb_sectors << BDRV_SECTOR_BITS;
         buf += len;
@@ -721,8 +725,8 @@ int bdrv_pread(BlockDriverState *bs, int64_t offset,
 
     /* add data from the last sector */
     if (count > 0) {
-        if (bdrv_read(bs, sector_num, tmp_buf, 1) < 0)
+        if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
-            return -EIO;
+            return ret;
         memcpy(buf, tmp_buf, count);
     }
     return count1;
@@ -734,6 +738,7 @@ int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
     uint8_t tmp_buf[BDRV_SECTOR_SIZE];
     int len, nb_sectors, count;
     int64_t sector_num;
+    int ret;
 
     count = count1;
     /* first write to align to sector start */
@@ -742,11 +747,11 @@ int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
         len = count;
     sector_num = offset >> BDRV_SECTOR_BITS;
     if (len > 0) {
-        if (bdrv_read(bs, sector_num, tmp_buf, 1) < 0)
+        if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
-            return -EIO;
+            return ret;
         memcpy(tmp_buf + (offset & (BDRV_SECTOR_SIZE - 1)), buf, len);
-        if (bdrv_write(bs, sector_num, tmp_buf, 1) < 0)
+        if ((ret = bdrv_write(bs, sector_num, tmp_buf, 1)) < 0)
-            return -EIO;
+            return ret;
         count -= len;
         if (count == 0)
             return count1;
@@ -757,8 +762,8 @@ int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
     /* write the sectors "in place" */
     nb_sectors = count >> BDRV_SECTOR_BITS;
     if (nb_sectors > 0) {
-        if (bdrv_write(bs, sector_num, buf, nb_sectors) < 0)
+        if ((ret = bdrv_write(bs, sector_num, buf, nb_sectors)) < 0)
-            return -EIO;
+            return ret;
         sector_num += nb_sectors;
         len = nb_sectors << BDRV_SECTOR_BITS;
         buf += len;
@@ -767,15 +772,52 @@ int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
 
     /* add data from the last sector */
     if (count > 0) {
-        if (bdrv_read(bs, sector_num, tmp_buf, 1) < 0)
+        if ((ret = bdrv_read(bs, sector_num, tmp_buf, 1)) < 0)
-            return -EIO;
+            return ret;
         memcpy(tmp_buf, buf, count);
-        if (bdrv_write(bs, sector_num, tmp_buf, 1) < 0)
+        if ((ret = bdrv_write(bs, sector_num, tmp_buf, 1)) < 0)
-            return -EIO;
+            return ret;
     }
     return count1;
 }
 
+/*
+ * Writes to the file and ensures that no writes are reordered across this
+ * request (acts as a barrier)
+ *
+ * Returns 0 on success, -errno in error cases.
+ */
+int bdrv_pwrite_sync(BlockDriverState *bs, int64_t offset,
+    const void *buf, int count)
+{
+    int ret;
+
+    ret = bdrv_pwrite(bs, offset, buf, count);
+    if (ret < 0) {
+        return ret;
+    }
+
+    /* No flush needed for cache=writethrough, it uses O_DSYNC */
+    if ((bs->open_flags & BDRV_O_CACHE_MASK) != 0) {
+        bdrv_flush(bs);
+    }
+
+    return 0;
+}
+
+/*
+ * Writes to the file and ensures that no writes are reordered across this
+ * request (acts as a barrier)
+ *
+ * Returns 0 on success, -errno in error cases.
+ */
+int bdrv_write_sync(BlockDriverState *bs, int64_t sector_num,
+    const uint8_t *buf, int nb_sectors)
+{
+    return bdrv_pwrite_sync(bs, BDRV_SECTOR_SIZE * sector_num,
+        buf, BDRV_SECTOR_SIZE * nb_sectors);
+}
+
 /**
  * Truncate file to 'offset' bytes (needed only for file protocols)
  */
@@ -1139,61 +1181,203 @@ int bdrv_is_allocated(BlockDriverState *bs, int64_t sector_num, int nb_sectors,
     return bs->drv->bdrv_is_allocated(bs, sector_num, nb_sectors, pnum);
 }
 
-void bdrv_info(Monitor *mon)
+static void bdrv_print_dict(QObject *obj, void *opaque)
 {
-    BlockDriverState *bs;
+    QDict *bs_dict;
+    Monitor *mon = opaque;
 
-    for (bs = bdrv_first; bs != NULL; bs = bs->next) {
+    bs_dict = qobject_to_qdict(obj);
-        monitor_printf(mon, "%s:", bs->device_name);
+
-        monitor_printf(mon, " type=");
+    monitor_printf(mon, "%s: type=%s removable=%d",
-        switch(bs->type) {
+                        qdict_get_str(bs_dict, "device"),
-        case BDRV_TYPE_HD:
+                        qdict_get_str(bs_dict, "type"),
-            monitor_printf(mon, "hd");
+                        qdict_get_bool(bs_dict, "removable"));
-            break;
+
-        case BDRV_TYPE_CDROM:
+    if (qdict_get_bool(bs_dict, "removable")) {
-            monitor_printf(mon, "cdrom");
+        monitor_printf(mon, " locked=%d", qdict_get_bool(bs_dict, "locked"));
-            break;
-        case BDRV_TYPE_FLOPPY:
-            monitor_printf(mon, "floppy");
-            break;
-        }
-        monitor_printf(mon, " removable=%d", bs->removable);
-        if (bs->removable) {
-            monitor_printf(mon, " locked=%d", bs->locked);
-        }
-        if (bs->drv) {
-            monitor_printf(mon, " file=");
-            monitor_print_filename(mon, bs->filename);
-            if (bs->backing_file[0] != '\0') {
-                monitor_printf(mon, " backing_file=");
-                monitor_print_filename(mon, bs->backing_file);
-            }
-            monitor_printf(mon, " ro=%d", bs->read_only);
-            monitor_printf(mon, " drv=%s", bs->drv->format_name);
-            monitor_printf(mon, " encrypted=%d", bdrv_is_encrypted(bs));
-        } else {
-            monitor_printf(mon, " [not inserted]");
-        }
-        monitor_printf(mon, "\n");
     }
+
+    if (qdict_haskey(bs_dict, "inserted")) {
+        QDict *qdict = qobject_to_qdict(qdict_get(bs_dict, "inserted"));
+
+        monitor_printf(mon, " file=");
+        monitor_print_filename(mon, qdict_get_str(qdict, "file"));
+        if (qdict_haskey(qdict, "backing_file")) {
+            monitor_printf(mon, " backing_file=");
+            monitor_print_filename(mon, qdict_get_str(qdict, "backing_file"));
+        }
+        monitor_printf(mon, " ro=%d drv=%s encrypted=%d",
+                            qdict_get_bool(qdict, "ro"),
+                            qdict_get_str(qdict, "drv"),
+                            qdict_get_bool(qdict, "encrypted"));
+    } else {
+        monitor_printf(mon, " [not inserted]");
+    }
+
+    monitor_printf(mon, "\n");
 }
 
-/* The "info blockstats" command. */
+void bdrv_info_print(Monitor *mon, const QObject *data)
-void bdrv_info_stats(Monitor *mon)
 {
+    qlist_iter(qobject_to_qlist(data), bdrv_print_dict, mon);
+}
+
+/**
+ * bdrv_info(): Block devices information
+ *
+ * Each block device information is stored in a QDict and the
+ * returned QObject is a QList of all devices.
+ *
+ * The QDict contains the following:
+ *
+ * - "device": device name
+ * - "type": device type
+ * - "removable": true if the device is removable, false otherwise
+ * - "locked": true if the device is locked, false otherwise
+ * - "inserted": only present if the device is inserted, it is a QDict
+ *    containing the following:
+ *          - "file": device file name
+ *          - "ro": true if read-only, false otherwise
+ *          - "drv": driver format name
+ *          - "backing_file": backing file name if one is used
+ *          - "encrypted": true if encrypted, false otherwise
+ *
+ * Example:
+ *
+ * [ { "device": "ide0-hd0", "type": "hd", "removable": false, "locked": false,
+ *     "inserted": { "file": "/tmp/foobar", "ro": false, "drv": "qcow2" } },
+ *   { "device": "floppy0", "type": "floppy", "removable": true,
+ *     "locked": false } ]
+ */
+void bdrv_info(Monitor *mon, QObject **ret_data)
+{
+    QList *bs_list;
     BlockDriverState *bs;
 
+    bs_list = qlist_new();
+
     for (bs = bdrv_first; bs != NULL; bs = bs->next) {
-        monitor_printf(mon, "%s:"
+        QObject *bs_obj;
-                       " rd_bytes=%" PRIu64
+        const char *type = "unknown";
-                       " wr_bytes=%" PRIu64
+
-                       " rd_operations=%" PRIu64
+        switch(bs->type) {
-                       " wr_operations=%" PRIu64
+        case BDRV_TYPE_HD:
-                       "\n",
+            type = "hd";
-                       bs->device_name,
+            break;
-                       bs->rd_bytes, bs->wr_bytes,
+        case BDRV_TYPE_CDROM:
-                       bs->rd_ops, bs->wr_ops);
+            type = "cdrom";
+            break;
+        case BDRV_TYPE_FLOPPY:
+            type = "floppy";
+            break;
+        }
+
+        bs_obj = qobject_from_jsonf("{ 'device': %s, 'type': %s, "
+                                    "'removable': %i, 'locked': %i }",
+                                    bs->device_name, type, bs->removable,
+                                    bs->locked);
+        assert(bs_obj != NULL);
+
+        if (bs->drv) {
+            QObject *obj;
+            QDict *bs_dict = qobject_to_qdict(bs_obj);
+
+            obj = qobject_from_jsonf("{ 'file': %s, 'ro': %i, 'drv': %s, "
+                                     "'encrypted': %i }",
+                                     bs->filename, bs->read_only,
+                                     bs->drv->format_name,
+                                     bdrv_is_encrypted(bs));
+            assert(obj != NULL);
+            if (bs->backing_file[0] != '\0') {
+                QDict *qdict = qobject_to_qdict(obj);
+                qdict_put(qdict, "backing_file",
+                          qstring_from_str(bs->backing_file));
+            }
+
+            qdict_put_obj(bs_dict, "inserted", obj);
+        }
+        qlist_append_obj(bs_list, bs_obj);
     }
+
+    *ret_data = QOBJECT(bs_list);
+}
+
+static void bdrv_stats_iter(QObject *data, void *opaque)
+{
+    QDict *qdict;
+    Monitor *mon = opaque;
+
+    qdict = qobject_to_qdict(data);
+    monitor_printf(mon, "%s:", qdict_get_str(qdict, "device"));
+
+    qdict = qobject_to_qdict(qdict_get(qdict, "stats"));
+    monitor_printf(mon, " rd_bytes=%" PRId64
+                        " wr_bytes=%" PRId64
+                        " rd_operations=%" PRId64
+                        " wr_operations=%" PRId64
+                        "\n",
+                        qdict_get_int(qdict, "rd_bytes"),
+                        qdict_get_int(qdict, "wr_bytes"),
+                        qdict_get_int(qdict, "rd_operations"),
+                        qdict_get_int(qdict, "wr_operations"));
+}
+
+void bdrv_stats_print(Monitor *mon, const QObject *data)
+{
+    qlist_iter(qobject_to_qlist(data), bdrv_stats_iter, mon);
+}
+
+/**
+ * bdrv_info_stats(): show block device statistics
+ *
+ * Each device statistic information is stored in a QDict and
+ * the returned QObject is a QList of all devices.
+ *
+ * The QDict contains the following:
+ *
+ * - "device": device name
+ * - "stats": A QDict with the statistics information, it contains:
+ *     - "rd_bytes": bytes read
+ *     - "wr_bytes": bytes written
+ *     - "rd_operations": read operations
+ *     - "wr_operations": write operations
+ * 
+ * Example:
+ *
+ * [ { "device": "ide0-hd0",
+ *               "stats": { "rd_bytes": 512,
+ *                          "wr_bytes": 0,
+ *                          "rd_operations": 1,
+ *                          "wr_operations": 0 } },
+ *   { "device": "ide1-cd0",
+ *               "stats": { "rd_bytes": 0,
+ *                          "wr_bytes": 0,
+ *                          "rd_operations": 0,
+ *                          "wr_operations": 0 } } ]
+ */
+void bdrv_info_stats(Monitor *mon, QObject **ret_data)
+{
+    QObject *obj;
+    QList *devices;
+    BlockDriverState *bs;
+
+    devices = qlist_new();
+
+    for (bs = bdrv_first; bs != NULL; bs = bs->next) {
+        obj = qobject_from_jsonf("{ 'device': %s, 'stats': {"
+                                 "'rd_bytes': %" PRId64 ","
+                                 "'wr_bytes': %" PRId64 ","
+                                 "'rd_operations': %" PRId64 ","
+                                 "'wr_operations': %" PRId64
+                                 "} }",
+                                 bs->device_name,
+                                 bs->rd_bytes, bs->wr_bytes,
+                                 bs->rd_ops, bs->wr_ops);
+        assert(obj != NULL);
+        qlist_append_obj(devices, obj);
+    }
+
+    *ret_data = QOBJECT(devices);
 }
 
 const char *bdrv_get_encrypted_filename(BlockDriverState *bs)
@@ -1463,8 +1647,11 @@ static void multiwrite_user_cb(MultiwriteCB *mcb)
 
     for (i = 0; i < mcb->num_callbacks; i++) {
         mcb->callbacks[i].cb(mcb->callbacks[i].opaque, mcb->error);
+        if (mcb->callbacks[i].free_qiov) {
+            qemu_iovec_destroy(mcb->callbacks[i].free_qiov);
+        }
         qemu_free(mcb->callbacks[i].free_qiov);
-        qemu_free(mcb->callbacks[i].free_buf);
+        qemu_vfree(mcb->callbacks[i].free_buf);
     }
 }
 
@@ -1472,23 +1659,32 @@ static void multiwrite_cb(void *opaque, int ret)
 {
     MultiwriteCB *mcb = opaque;
 
-    if (ret < 0) {
+    if (ret < 0 && !mcb->error) {
         mcb->error = ret;
-        multiwrite_user_cb(mcb);
     }
 
     mcb->num_requests--;
     if (mcb->num_requests == 0) {
-        if (mcb->error == 0) {
+        multiwrite_user_cb(mcb);
-            multiwrite_user_cb(mcb);
-        }
         qemu_free(mcb);
     }
 }
 
 static int multiwrite_req_compare(const void *a, const void *b)
 {
-    return (((BlockRequest*) a)->sector - ((BlockRequest*) b)->sector);
+    const BlockRequest *req1 = a, *req2 = b;
+
+    /*
+     * Note that we can't simply subtract req2->sector from req1->sector
+     * here as that could overflow the return value.
+     */
+    if (req1->sector > req2->sector) {
+        return 1;
+    } else if (req1->sector < req2->sector) {
+        return -1;
+    } else {
+        return 0;
+    }
 }
 
 /*
@@ -1524,6 +1720,10 @@ static int multiwrite_merge(BlockDriverState *bs, BlockRequest *reqs,
             merge = bs->drv->bdrv_merge_requests(bs, &reqs[outidx], &reqs[i]);
         }
 
+        if (reqs[outidx].qiov->niov + reqs[i].qiov->niov + 1 > IOV_MAX) {
+            merge = 0;
+        }
+
         if (merge) {
             size_t size;
             QEMUIOVector *qiov = qemu_mallocz(sizeof(*qiov));
@@ -1547,7 +1747,7 @@ static int multiwrite_merge(BlockDriverState *bs, BlockRequest *reqs,
             // Add the second request
             qemu_iovec_concat(qiov, reqs[i].qiov, reqs[i].qiov->size);
 
-            reqs[outidx].nb_sectors += reqs[i].nb_sectors;
+            reqs[outidx].nb_sectors = qiov->size >> 9;
             reqs[outidx].qiov = qiov;
 
             mcb->callbacks[i].free_qiov = reqs[outidx].qiov;
@@ -1599,8 +1799,29 @@ int bdrv_aio_multiwrite(BlockDriverState *bs, BlockRequest *reqs, int num_reqs)
     // Check for mergable requests
     num_reqs = multiwrite_merge(bs, reqs, num_reqs, mcb);
 
-    // Run the aio requests
+    /*
+     * Run the aio requests. As soon as one request can't be submitted
+     * successfully, fail all requests that are not yet submitted (we must
+     * return failure for all requests anyway)
+     *
+     * num_requests cannot be set to the right value immediately: If
+     * bdrv_aio_writev fails for some request, num_requests would be too high
+     * and therefore multiwrite_cb() would never recognize the multiwrite
+     * request as completed. We also cannot use the loop variable i to set it
+     * when the first request fails because the callback may already have been
+     * called for previously submitted requests. Thus, num_requests must be
+     * incremented for each request that is submitted.
+     *
+     * The problem that callbacks may be called early also means that we need
+     * to take care that num_requests doesn't become 0 before all requests are
+     * submitted - multiwrite_cb() would consider the multiwrite request
+     * completed. A dummy request that is "completed" by a manual call to
+     * multiwrite_cb() takes care of this.
+     */
+    mcb->num_requests = 1;
+
     for (i = 0; i < num_reqs; i++) {
+        mcb->num_requests++;
         acb = bdrv_aio_writev(bs, reqs[i].sector, reqs[i].qiov,
             reqs[i].nb_sectors, multiwrite_cb, mcb);
 
@@ -1608,22 +1829,25 @@ int bdrv_aio_multiwrite(BlockDriverState *bs, BlockRequest *reqs, int num_reqs)
             // We can only fail the whole thing if no request has been
             // submitted yet. Otherwise we'll wait for the submitted AIOs to
             // complete and report the error in the callback.
-            if (mcb->num_requests == 0) {
+            if (i == 0) {
-                reqs[i].error = EIO;
                 goto fail;
             } else {
-                mcb->error = EIO;
+                multiwrite_cb(mcb, -EIO);
                 break;
             }
-        } else {
-            mcb->num_requests++;
         }
     }
 
+    /* Complete the dummy request */
+    multiwrite_cb(mcb, 0);
+
     return 0;
 
 fail:
-    free(mcb);
+    for (i = 0; i < mcb->num_callbacks; i++) {
+        reqs[i].error = -EIO;
+    }
+    qemu_free(mcb);
     return -1;
 }
 

block.h

@@ -4,6 +4,7 @@
 #include "qemu-aio.h"
 #include "qemu-common.h"
 #include "qemu-option.h"
+#include "qobject.h"
 
 /* block.c */
 typedef struct BlockDriver BlockDriver;
@@ -45,8 +46,10 @@ typedef struct QEMUSnapshotInfo {
 #define BDRV_SECTOR_SIZE   (1 << BDRV_SECTOR_BITS)
 #define BDRV_SECTOR_MASK   ~(BDRV_SECTOR_SIZE - 1);
 
-void bdrv_info(Monitor *mon);
+void bdrv_info_print(Monitor *mon, const QObject *data);
-void bdrv_info_stats(Monitor *mon);
+void bdrv_info(Monitor *mon, QObject **ret_data);
+void bdrv_stats_print(Monitor *mon, const QObject *data);
+void bdrv_info_stats(Monitor *mon, QObject **ret_data);
 
 void bdrv_init(void);
 void bdrv_init_with_whitelist(void);
@@ -74,6 +77,10 @@ int bdrv_pread(BlockDriverState *bs, int64_t offset,
                void *buf, int count);
 int bdrv_pwrite(BlockDriverState *bs, int64_t offset,
                 const void *buf, int count);
+int bdrv_pwrite_sync(BlockDriverState *bs, int64_t offset,
+    const void *buf, int count);
+int bdrv_write_sync(BlockDriverState *bs, int64_t sector_num,
+    const uint8_t *buf, int nb_sectors);
 int bdrv_truncate(BlockDriverState *bs, int64_t offset);
 int64_t bdrv_getlength(BlockDriverState *bs);
 void bdrv_get_geometry(BlockDriverState *bs, uint64_t *nb_sectors_ptr);

block/curl.c

@@ -309,7 +309,7 @@ static int curl_open(BlockDriverState *bs, const char *filename, int flags)
 
     static int inited = 0;
 
-    file = strdup(filename);
+    file = qemu_strdup(filename);
     s->readahead_size = READ_AHEAD_SIZE;
 
     /* Parse a trailing ":readahead=#:" param, if present. */

block/dmg.c

@@ -90,24 +90,21 @@ static int dmg_open(BlockDriverState *bs, const char *filename, int flags)
 
     /* read offset of info blocks */
     if(lseek(s->fd,-0x1d8,SEEK_END)<0) {
-dmg_close:
+        goto fail;
-	close(s->fd);
-	/* open raw instead */
-	bs->drv=bdrv_find_format("raw");
-	return bs->drv->bdrv_open(bs, filename, flags);
     }
+
     info_begin=read_off(s->fd);
     if(info_begin==0)
-	goto dmg_close;
+	goto fail;
     if(lseek(s->fd,info_begin,SEEK_SET)<0)
-	goto dmg_close;
+	goto fail;
     if(read_uint32(s->fd)!=0x100)
-	goto dmg_close;
+	goto fail;
     if((count = read_uint32(s->fd))==0)
-	goto dmg_close;
+	goto fail;
     info_end = info_begin+count;
     if(lseek(s->fd,0xf8,SEEK_CUR)<0)
-	goto dmg_close;
+	goto fail;
 
     /* read offsets */
     last_in_offset = last_out_offset = 0;
@@ -116,14 +113,14 @@ dmg_close:
 
 	count = read_uint32(s->fd);
 	if(count==0)
-	    goto dmg_close;
+	    goto fail;
 	type = read_uint32(s->fd);
 	if(type!=0x6d697368 || count<244)
 	    lseek(s->fd,count-4,SEEK_CUR);
 	else {
 	    int new_size, chunk_count;
 	    if(lseek(s->fd,200,SEEK_CUR)<0)
-	        goto dmg_close;
+	        goto fail;
 	    chunk_count = (count-204)/40;
 	    new_size = sizeof(uint64_t) * (s->n_chunks + chunk_count);
 	    s->types = qemu_realloc(s->types, new_size/2);
@@ -142,7 +139,7 @@ dmg_close:
 		    chunk_count--;
 		    i--;
 		    if(lseek(s->fd,36,SEEK_CUR)<0)
-			goto dmg_close;
+			goto fail;
 		    continue;
 		}
 		read_uint32(s->fd);
@@ -163,11 +160,14 @@ dmg_close:
     s->compressed_chunk = qemu_malloc(max_compressed_size+1);
     s->uncompressed_chunk = qemu_malloc(512*max_sectors_per_chunk);
     if(inflateInit(&s->zstream) != Z_OK)
-	goto dmg_close;
+	goto fail;
 
     s->current_chunk = s->n_chunks;
 
     return 0;
+fail:
+    close(s->fd);
+    return -1;
 }
 
 static inline int is_sector_in_chunk(BDRVDMGState* s,

block/qcow.c

@@ -277,8 +277,9 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
         /* update the L1 entry */
         s->l1_table[l1_index] = l2_offset;
         tmp = cpu_to_be64(l2_offset);
-        if (bdrv_pwrite(s->hd, s->l1_table_offset + l1_index * sizeof(tmp),
+        if (bdrv_pwrite_sync(s->hd,
-                        &tmp, sizeof(tmp)) != sizeof(tmp))
+                s->l1_table_offset + l1_index * sizeof(tmp),
+                &tmp, sizeof(tmp)) < 0)
             return 0;
         new_l2_table = 1;
     }
@@ -306,8 +307,8 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
     l2_table = s->l2_cache + (min_index << s->l2_bits);
     if (new_l2_table) {
         memset(l2_table, 0, s->l2_size * sizeof(uint64_t));
-        if (bdrv_pwrite(s->hd, l2_offset, l2_table, s->l2_size * sizeof(uint64_t)) !=
+        if (bdrv_pwrite_sync(s->hd, l2_offset, l2_table,
-            s->l2_size * sizeof(uint64_t))
+                s->l2_size * sizeof(uint64_t)) < 0)
             return 0;
     } else {
         if (bdrv_pread(s->hd, l2_offset, l2_table, s->l2_size * sizeof(uint64_t)) !=
@@ -372,8 +373,8 @@ static uint64_t get_cluster_offset(BlockDriverState *bs,
         /* update L2 table */
         tmp = cpu_to_be64(cluster_offset);
         l2_table[l2_index] = tmp;
-        if (bdrv_pwrite(s->hd,
+        if (bdrv_pwrite_sync(s->hd, l2_offset + l2_index * sizeof(tmp),
-                        l2_offset + l2_index * sizeof(tmp), &tmp, sizeof(tmp)) != sizeof(tmp))
+                &tmp, sizeof(tmp)) < 0)
             return 0;
     }
     return cluster_offset;
@@ -821,8 +822,9 @@ static int qcow_make_empty(BlockDriverState *bs)
     int ret;
 
     memset(s->l1_table, 0, l1_length);
-    if (bdrv_pwrite(s->hd, s->l1_table_offset, s->l1_table, l1_length) < 0)
+    if (bdrv_pwrite_sync(s->hd, s->l1_table_offset, s->l1_table,
-	return -1;
+            l1_length) < 0)
+        return -1;
     ret = bdrv_truncate(s->hd, s->l1_table_offset + l1_length);
     if (ret < 0)
         return ret;

block/qcow2-cluster.c

@@ -33,7 +33,7 @@ int qcow2_grow_l1_table(BlockDriverState *bs, int min_size)
     BDRVQcowState *s = bs->opaque;
     int new_l1_size, new_l1_size2, ret, i;
     uint64_t *new_l1_table;
-    uint64_t new_l1_table_offset;
+    int64_t new_l1_table_offset;
     uint8_t data[12];
 
     new_l1_size = s->l1_size;
@@ -55,11 +55,15 @@ int qcow2_grow_l1_table(BlockDriverState *bs, int min_size)
 
     /* write new table (align to cluster) */
     new_l1_table_offset = qcow2_alloc_clusters(bs, new_l1_size2);
+    if (new_l1_table_offset < 0) {
+        qemu_free(new_l1_table);
+        return new_l1_table_offset;
+    }
 
     for(i = 0; i < s->l1_size; i++)
         new_l1_table[i] = cpu_to_be64(new_l1_table[i]);
-    ret = bdrv_pwrite(s->hd, new_l1_table_offset, new_l1_table, new_l1_size2);
+    ret = bdrv_pwrite_sync(s->hd, new_l1_table_offset, new_l1_table, new_l1_size2);
-    if (ret != new_l1_size2)
+    if (ret < 0)
         goto fail;
     for(i = 0; i < s->l1_size; i++)
         new_l1_table[i] = be64_to_cpu(new_l1_table[i]);
@@ -67,9 +71,10 @@ int qcow2_grow_l1_table(BlockDriverState *bs, int min_size)
     /* set new table */
     cpu_to_be32w((uint32_t*)data, new_l1_size);
     cpu_to_be64w((uint64_t*)(data + 4), new_l1_table_offset);
-    if (bdrv_pwrite(s->hd, offsetof(QCowHeader, l1_size), data,
+    ret = bdrv_pwrite_sync(s->hd, offsetof(QCowHeader, l1_size), data,sizeof(data));
-                sizeof(data)) != sizeof(data))
+    if (ret < 0) {
         goto fail;
+    }
     qemu_free(s->l1_table);
     qcow2_free_clusters(bs, s->l1_table_offset, s->l1_size * sizeof(uint64_t));
     s->l1_table_offset = new_l1_table_offset;
@@ -77,8 +82,9 @@ int qcow2_grow_l1_table(BlockDriverState *bs, int min_size)
     s->l1_size = new_l1_size;
     return 0;
  fail:
-    qemu_free(s->l1_table);
+    qemu_free(new_l1_table);
-    return -EIO;
+    qcow2_free_clusters(bs, new_l1_table_offset, new_l1_size2);
+    return ret;
 }
 
 void qcow2_l2_cache_reset(BlockDriverState *bs)
@@ -182,17 +188,17 @@ static int write_l1_entry(BDRVQcowState *s, int l1_index)
 {
     uint64_t buf[L1_ENTRIES_PER_SECTOR];
     int l1_start_index;
-    int i;
+    int i, ret;
 
     l1_start_index = l1_index & ~(L1_ENTRIES_PER_SECTOR - 1);
     for (i = 0; i < L1_ENTRIES_PER_SECTOR; i++) {
         buf[i] = cpu_to_be64(s->l1_table[l1_start_index + i]);
     }
 
-    if (bdrv_pwrite(s->hd, s->l1_table_offset + 8 * l1_start_index,
+    ret = bdrv_pwrite_sync(s->hd, s->l1_table_offset + 8 * l1_start_index,
-        buf, sizeof(buf)) != sizeof(buf))
+        buf, sizeof(buf));
-    {
+    if (ret < 0) {
-        return -1;
+        return ret;
     }
 
     return 0;
@@ -213,18 +219,16 @@ static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
     BDRVQcowState *s = bs->opaque;
     int min_index;
     uint64_t old_l2_offset;
-    uint64_t *l2_table, l2_offset;
+    uint64_t *l2_table;
+    int64_t l2_offset;
+    int ret;
 
     old_l2_offset = s->l1_table[l1_index];
 
     /* allocate a new l2 entry */
 
     l2_offset = qcow2_alloc_clusters(bs, s->l2_size * sizeof(uint64_t));
-
+    if (l2_offset < 0) {
-    /* update the L1 entry */
-
-    s->l1_table[l1_index] = l2_offset | QCOW_OFLAG_COPIED;
-    if (write_l1_entry(s, l1_index) < 0) {
         return NULL;
     }
 
@@ -241,13 +245,20 @@ static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
         if (bdrv_pread(s->hd, old_l2_offset,
                        l2_table, s->l2_size * sizeof(uint64_t)) !=
             s->l2_size * sizeof(uint64_t))
-            return NULL;
+            goto fail;
     }
     /* write the l2 table to the file */
-    if (bdrv_pwrite(s->hd, l2_offset,
+    ret = bdrv_pwrite_sync(s->hd, l2_offset, l2_table,
-                    l2_table, s->l2_size * sizeof(uint64_t)) !=
+        s->l2_size * sizeof(uint64_t));
-        s->l2_size * sizeof(uint64_t))
+    if (ret < 0) {
-        return NULL;
+        goto fail;
+    }
+
+    /* update the L1 entry */
+    s->l1_table[l1_index] = l2_offset | QCOW_OFLAG_COPIED;
+    if (write_l1_entry(s, l1_index) < 0) {
+        goto fail;
+    }
 
     /* update the l2 cache entry */
 
@@ -255,6 +266,11 @@ static uint64_t *l2_allocate(BlockDriverState *bs, int l1_index)
     s->l2_cache_counts[min_index] = 1;
 
     return l2_table;
+
+fail:
+    s->l1_table[l1_index] = old_l2_offset;
+    qcow2_l2_cache_reset(bs);
+    return NULL;
 }
 
 static int count_contiguous_clusters(uint64_t nb_clusters, int cluster_size,
@@ -370,8 +386,8 @@ static int copy_sectors(BlockDriverState *bs, uint64_t start_sect,
                         s->cluster_data, n, 1,
                         &s->aes_encrypt_key);
     }
-    ret = bdrv_write(s->hd, (cluster_offset >> 9) + n_start,
+    ret = bdrv_write_sync(s->hd, (cluster_offset >> 9) + n_start,
-                     s->cluster_data, n);
+        s->cluster_data, n);
     if (ret < 0)
         return ret;
     return 0;
@@ -479,8 +495,8 @@ out:
  * the l2 table offset in the qcow2 file and the cluster index
  * in the l2 table are given to the caller.
  *
+ * Returns 0 on success, -errno in failure case
  */
-
 static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
                              uint64_t **new_l2_table,
                              uint64_t *new_l2_offset,
@@ -496,8 +512,9 @@ static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
     l1_index = offset >> (s->l2_bits + s->cluster_bits);
     if (l1_index >= s->l1_size) {
         ret = qcow2_grow_l1_table(bs, l1_index + 1);
-        if (ret < 0)
+        if (ret < 0) {
-            return 0;
+            return ret;
+        }
     }
     l2_offset = s->l1_table[l1_index];
 
@@ -507,14 +524,16 @@ static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
         /* load the l2 table in memory */
         l2_offset &= ~QCOW_OFLAG_COPIED;
         l2_table = l2_load(bs, l2_offset);
-        if (l2_table == NULL)
+        if (l2_table == NULL) {
-            return 0;
+            return -EIO;
+        }
     } else {
         if (l2_offset)
             qcow2_free_clusters(bs, l2_offset, s->l2_size * sizeof(uint64_t));
         l2_table = l2_allocate(bs, l1_index);
-        if (l2_table == NULL)
+        if (l2_table == NULL) {
-            return 0;
+            return -EIO;
+        }
         l2_offset = s->l1_table[l1_index] & ~QCOW_OFLAG_COPIED;
     }
 
@@ -526,7 +545,7 @@ static int get_cluster_table(BlockDriverState *bs, uint64_t offset,
     *new_l2_offset = l2_offset;
     *new_l2_index = l2_index;
 
-    return 1;
+    return 0;
 }
 
 /*
@@ -548,12 +567,14 @@ uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
 {
     BDRVQcowState *s = bs->opaque;
     int l2_index, ret;
-    uint64_t l2_offset, *l2_table, cluster_offset;
+    uint64_t l2_offset, *l2_table;
+    int64_t cluster_offset;
     int nb_csectors;
 
     ret = get_cluster_table(bs, offset, &l2_table, &l2_offset, &l2_index);
-    if (ret == 0)
+    if (ret < 0) {
         return 0;
+    }
 
     cluster_offset = be64_to_cpu(l2_table[l2_index]);
     if (cluster_offset & QCOW_OFLAG_COPIED)
@@ -563,6 +584,10 @@ uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
         qcow2_free_any_clusters(bs, cluster_offset, 1);
 
     cluster_offset = qcow2_alloc_bytes(bs, compressed_size);
+    if (cluster_offset < 0) {
+        return 0;
+    }
+
     nb_csectors = ((cluster_offset + compressed_size - 1) >> 9) -
                   (cluster_offset >> 9);
 
@@ -574,10 +599,10 @@ uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
     /* compressed clusters never have the copied flag */
 
     l2_table[l2_index] = cpu_to_be64(cluster_offset);
-    if (bdrv_pwrite(s->hd,
+    if (bdrv_pwrite_sync(s->hd,
                     l2_offset + l2_index * sizeof(uint64_t),
                     l2_table + l2_index,
-                    sizeof(uint64_t)) != sizeof(uint64_t))
+                    sizeof(uint64_t)) < 0)
         return 0;
 
     return cluster_offset;
@@ -595,22 +620,23 @@ static int write_l2_entries(BDRVQcowState *s, uint64_t *l2_table,
     int start_offset = (8 * l2_index) & ~511;
     int end_offset = (8 * (l2_index + num) + 511) & ~511;
     size_t len = end_offset - start_offset;
+    int ret;
 
-    if (bdrv_pwrite(s->hd, l2_offset + start_offset, &l2_table[l2_start_index],
+    ret = bdrv_pwrite_sync(s->hd, l2_offset + start_offset,
-        len) != len)
+        &l2_table[l2_start_index], len);
-    {
+    if (ret < 0) {
-        return -1;
+        return ret;
     }
 
     return 0;
 }
 
-int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, uint64_t cluster_offset,
+int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m)
-    QCowL2Meta *m)
 {
     BDRVQcowState *s = bs->opaque;
     int i, j = 0, l2_index, ret;
     uint64_t *old_cluster, start_sect, l2_offset, *l2_table;
+    uint64_t cluster_offset = m->cluster_offset;
 
     if (m->nb_clusters == 0)
         return 0;
@@ -633,10 +659,11 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, uint64_t cluster_offset,
             goto err;
     }
 
-    ret = -EIO;
     /* update L2 table */
-    if (!get_cluster_table(bs, m->offset, &l2_table, &l2_offset, &l2_index))
+    ret = get_cluster_table(bs, m->offset, &l2_table, &l2_offset, &l2_index);
+    if (ret < 0) {
         goto err;
+    }
 
     for (i = 0; i < m->nb_clusters; i++) {
         /* if two concurrent writes happen to the same unallocated cluster
@@ -652,8 +679,9 @@ int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, uint64_t cluster_offset,
                     (i << s->cluster_bits)) | QCOW_OFLAG_COPIED);
      }
 
-    if (write_l2_entries(s, l2_table, l2_offset, l2_index, m->nb_clusters) < 0) {
+    ret = write_l2_entries(s, l2_table, l2_offset, l2_index, m->nb_clusters);
-        ret = -1;
+    if (ret < 0) {
+        qcow2_l2_cache_reset(bs);
         goto err;
     }
 
@@ -670,30 +698,36 @@ err:
 /*
  * alloc_cluster_offset
  *
- * For a given offset of the disk image, return cluster offset in
+ * For a given offset of the disk image, return cluster offset in qcow2 file.
- * qcow2 file.
- *
  * If the offset is not found, allocate a new cluster.
  *
- * Return the cluster offset if successful,
+ * If the cluster was already allocated, m->nb_clusters is set to 0,
- * Return 0, otherwise.
+ * m->depends_on is set to NULL and the other fields in m are meaningless.
  *
+ * If the cluster is newly allocated, m->nb_clusters is set to the number of
+ * contiguous clusters that have been allocated. This may be 0 if the request
+ * conflict with another write request in flight; in this case, m->depends_on
+ * is set and the remaining fields of m are meaningless.
+ *
+ * If m->nb_clusters is non-zero, the other fields of m are valid and contain
+ * information about the first allocated cluster.
+ *
+ * Return 0 on success and -errno in error cases
  */
-
+int qcow2_alloc_cluster_offset(BlockDriverState *bs, uint64_t offset,
-uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
+    int n_start, int n_end, int *num, QCowL2Meta *m)
-                                    uint64_t offset,
-                                    int n_start, int n_end,
-                                    int *num, QCowL2Meta *m)
 {
     BDRVQcowState *s = bs->opaque;
     int l2_index, ret;
-    uint64_t l2_offset, *l2_table, cluster_offset;
+    uint64_t l2_offset, *l2_table;
+    int64_t cluster_offset;
     unsigned int nb_clusters, i = 0;
     QCowL2Meta *old_alloc;
 
     ret = get_cluster_table(bs, offset, &l2_table, &l2_offset, &l2_index);
-    if (ret == 0)
+    if (ret < 0) {
-        return 0;
+        return ret;
+    }
 
     nb_clusters = size_to_clusters(s, n_end << 9);
 
@@ -709,6 +743,7 @@ uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
 
         cluster_offset &= ~QCOW_OFLAG_COPIED;
         m->nb_clusters = 0;
+        m->depends_on = NULL;
 
         goto out;
     }
@@ -723,12 +758,15 @@ uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
     while (i < nb_clusters) {
         i += count_contiguous_clusters(nb_clusters - i, s->cluster_size,
                 &l2_table[l2_index], i, 0);
-
+        if ((i >= nb_clusters) || be64_to_cpu(l2_table[l2_index + i])) {
-        if(be64_to_cpu(l2_table[l2_index + i]))
             break;
+        }
 
         i += count_contiguous_free_clusters(nb_clusters - i,
                 &l2_table[l2_index + i]);
+        if (i >= nb_clusters) {
+            break;
+        }
 
         cluster_offset = be64_to_cpu(l2_table[l2_index + i]);
 
@@ -736,6 +774,7 @@ uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
                 (cluster_offset & QCOW_OFLAG_COMPRESSED))
             break;
     }
+    assert(i <= nb_clusters);
     nb_clusters = i;
 
     /*
@@ -779,6 +818,10 @@ uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
     /* allocate a new cluster */
 
     cluster_offset = qcow2_alloc_clusters(bs, nb_clusters * s->cluster_size);
+    if (cluster_offset < 0) {
+        QLIST_REMOVE(m, next_in_flight);
+        return cluster_offset;
+    }
 
     /* save info needed for meta data update */
     m->offset = offset;
@@ -787,10 +830,11 @@ uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
 
 out:
     m->nb_available = MIN(nb_clusters << (s->cluster_bits - 9), n_end);
+    m->cluster_offset = cluster_offset;
 
     *num = m->nb_available - n_start;
 
-    return cluster_offset;
+    return 0;
 }
 
 static int decompress_buffer(uint8_t *out_buf, int out_buf_size,

block/qcow2-refcount.c

@@ -27,7 +27,7 @@
 #include "block/qcow2.h"
 
 static int64_t alloc_clusters_noref(BlockDriverState *bs, int64_t size);
-static int update_refcount(BlockDriverState *bs,
+static int QEMU_WARN_UNUSED_RESULT update_refcount(BlockDriverState *bs,
                             int64_t offset, int64_t length,
                             int addend);
 
@@ -42,8 +42,8 @@ static int write_refcount_block(BDRVQcowState *s)
         return 0;
     }
 
-    if (bdrv_pwrite(s->hd, s->refcount_block_cache_offset,
+    if (bdrv_pwrite_sync(s->hd, s->refcount_block_cache_offset,
-            s->refcount_block_cache, size) != size)
+            s->refcount_block_cache, size) < 0)
     {
         return -EIO;
     }
@@ -123,121 +123,273 @@ static int get_refcount(BlockDriverState *bs, int64_t cluster_index)
     return be16_to_cpu(s->refcount_block_cache[block_index]);
 }
 
-static int grow_refcount_table(BlockDriverState *bs, int min_size)
+/*
+ * Rounds the refcount table size up to avoid growing the table for each single
+ * refcount block that is allocated.
+ */
+static unsigned int next_refcount_table_size(BDRVQcowState *s,
+    unsigned int min_size)
 {
-    BDRVQcowState *s = bs->opaque;
+    unsigned int min_clusters = (min_size >> (s->cluster_bits - 3)) + 1;
-    int new_table_size, new_table_size2, refcount_table_clusters, i, ret;
+    unsigned int refcount_table_clusters =
-    uint64_t *new_table;
+        MAX(1, s->refcount_table_size >> (s->cluster_bits - 3));
-    int64_t table_offset;
-    uint8_t data[12];
-    int old_table_size;
-    int64_t old_table_offset;
 
-    if (min_size <= s->refcount_table_size)
+    while (min_clusters > refcount_table_clusters) {
-        return 0;
+        refcount_table_clusters = (refcount_table_clusters * 3 + 1) / 2;
-    /* compute new table size */
-    refcount_table_clusters = s->refcount_table_size >> (s->cluster_bits - 3);
-    for(;;) {
-        if (refcount_table_clusters == 0) {
-            refcount_table_clusters = 1;
-        } else {
-            refcount_table_clusters = (refcount_table_clusters * 3 + 1) / 2;
-        }
-        new_table_size = refcount_table_clusters << (s->cluster_bits - 3);
-        if (min_size <= new_table_size)
-            break;
     }
-#ifdef DEBUG_ALLOC2
-    printf("grow_refcount_table from %d to %d\n",
-           s->refcount_table_size,
-           new_table_size);
-#endif
-    new_table_size2 = new_table_size * sizeof(uint64_t);
-    new_table = qemu_mallocz(new_table_size2);
-    memcpy(new_table, s->refcount_table,
-           s->refcount_table_size * sizeof(uint64_t));
-    for(i = 0; i < s->refcount_table_size; i++)
-        cpu_to_be64s(&new_table[i]);
-    /* Note: we cannot update the refcount now to avoid recursion */
-    table_offset = alloc_clusters_noref(bs, new_table_size2);
-    ret = bdrv_pwrite(s->hd, table_offset, new_table, new_table_size2);
-    if (ret != new_table_size2)
-        goto fail;
-    for(i = 0; i < s->refcount_table_size; i++)
-        be64_to_cpus(&new_table[i]);
 
-    cpu_to_be64w((uint64_t*)data, table_offset);
+    return refcount_table_clusters << (s->cluster_bits - 3);
-    cpu_to_be32w((uint32_t*)(data + 8), refcount_table_clusters);
-    if (bdrv_pwrite(s->hd, offsetof(QCowHeader, refcount_table_offset),
-                    data, sizeof(data)) != sizeof(data))
-        goto fail;
-    qemu_free(s->refcount_table);
-    old_table_offset = s->refcount_table_offset;
-    old_table_size = s->refcount_table_size;
-    s->refcount_table = new_table;
-    s->refcount_table_size = new_table_size;
-    s->refcount_table_offset = table_offset;
-
-    update_refcount(bs, table_offset, new_table_size2, 1);
-    qcow2_free_clusters(bs, old_table_offset, old_table_size * sizeof(uint64_t));
-    return 0;
- fail:
-    qemu_free(new_table);
-    return -EIO;
 }
 
 
+/* Checks if two offsets are described by the same refcount block */
+static int in_same_refcount_block(BDRVQcowState *s, uint64_t offset_a,
+    uint64_t offset_b)
+{
+    uint64_t block_a = offset_a >> (2 * s->cluster_bits - REFCOUNT_SHIFT);
+    uint64_t block_b = offset_b >> (2 * s->cluster_bits - REFCOUNT_SHIFT);
+
+    return (block_a == block_b);
+}
+
+/*
+ * Loads a refcount block. If it doesn't exist yet, it is allocated first
+ * (including growing the refcount table if needed).
+ *
+ * Returns the offset of the refcount block on success or -errno in error case
+ */
 static int64_t alloc_refcount_block(BlockDriverState *bs, int64_t cluster_index)
 {
     BDRVQcowState *s = bs->opaque;
-    int64_t offset, refcount_block_offset;
     unsigned int refcount_table_index;
     int ret;
-    uint64_t data64;
-    int cache = cache_refcount_updates;
 
-    /* Find L1 index and grow refcount table if needed */
+    /* Find the refcount block for the given cluster */
     refcount_table_index = cluster_index >> (s->cluster_bits - REFCOUNT_SHIFT);
-    if (refcount_table_index >= s->refcount_table_size) {
+
-        ret = grow_refcount_table(bs, refcount_table_index + 1);
+    if (refcount_table_index < s->refcount_table_size) {
-        if (ret < 0)
+
+        uint64_t refcount_block_offset =
+            s->refcount_table[refcount_table_index];
+
+        /* If it's already there, we're done */
+        if (refcount_block_offset) {
+            if (refcount_block_offset != s->refcount_block_cache_offset) {
+                ret = load_refcount_block(bs, refcount_block_offset);
+                if (ret < 0) {
+                    return ret;
+                }
+            }
+            return refcount_block_offset;
+        }
+    }
+
+    /*
+     * If we came here, we need to allocate something. Something is at least
+     * a cluster for the new refcount block. It may also include a new refcount
+     * table if the old refcount table is too small.
+     *
+     * Note that allocating clusters here needs some special care:
+     *
+     * - We can't use the normal qcow2_alloc_clusters(), it would try to
+     *   increase the refcount and very likely we would end up with an endless
+     *   recursion. Instead we must place the refcount blocks in a way that
+     *   they can describe them themselves.
+     *
+     * - We need to consider that at this point we are inside update_refcounts
+     *   and doing the initial refcount increase. This means that some clusters
+     *   have already been allocated by the caller, but their refcount isn't
+     *   accurate yet. free_cluster_index tells us where this allocation ends
+     *   as long as we don't overwrite it by freeing clusters.
+     *
+     * - alloc_clusters_noref and qcow2_free_clusters may load a different
+     *   refcount block into the cache
+     */
+
+    if (cache_refcount_updates) {
+        ret = write_refcount_block(s);
+        if (ret < 0) {
             return ret;
+        }
     }
 
-    /* Load or allocate the refcount block */
+    /* Allocate the refcount block itself and mark it as used */
-    refcount_block_offset = s->refcount_table[refcount_table_index];
+    uint64_t new_block = alloc_clusters_noref(bs, s->cluster_size);
-    if (!refcount_block_offset) {
+
-        if (cache_refcount_updates) {
+#ifdef DEBUG_ALLOC2
-            write_refcount_block(s);
+    fprintf(stderr, "qcow2: Allocate refcount block %d for %" PRIx64
-            cache_refcount_updates = 0;
+        " at %" PRIx64 "\n",
-        }
+        refcount_table_index, cluster_index << s->cluster_bits, new_block);
-        /* create a new refcount block */
+#endif
-        /* Note: we cannot update the refcount now to avoid recursion */
+
-        offset = alloc_clusters_noref(bs, s->cluster_size);
+    if (in_same_refcount_block(s, new_block, cluster_index << s->cluster_bits)) {
+        /* Zero the new refcount block before updating it */
         memset(s->refcount_block_cache, 0, s->cluster_size);
-        ret = bdrv_pwrite(s->hd, offset, s->refcount_block_cache, s->cluster_size);
+        s->refcount_block_cache_offset = new_block;
-        if (ret != s->cluster_size)
-            return -EINVAL;
-        s->refcount_table[refcount_table_index] = offset;
-        data64 = cpu_to_be64(offset);
-        ret = bdrv_pwrite(s->hd, s->refcount_table_offset +
-                          refcount_table_index * sizeof(uint64_t),
-                          &data64, sizeof(data64));
-        if (ret != sizeof(data64))
-            return -EINVAL;
 
-        refcount_block_offset = offset;
+        /* The block describes itself, need to update the cache */
-        s->refcount_block_cache_offset = offset;
+        int block_index = (new_block >> s->cluster_bits) &
-        update_refcount(bs, offset, s->cluster_size, 1);
+            ((1 << (s->cluster_bits - REFCOUNT_SHIFT)) - 1);
-        cache_refcount_updates = cache;
+        s->refcount_block_cache[block_index] = cpu_to_be16(1);
     } else {
-        if (refcount_block_offset != s->refcount_block_cache_offset) {
+        /* Described somewhere else. This can recurse at most twice before we
-            if (load_refcount_block(bs, refcount_block_offset) < 0)
+         * arrive at a block that describes itself. */
-                return -EIO;
+        ret = update_refcount(bs, new_block, s->cluster_size, 1);
+        if (ret < 0) {
+            goto fail_block;
         }
+
+        /* Initialize the new refcount block only after updating its refcount,
+         * update_refcount uses the refcount cache itself */
+        memset(s->refcount_block_cache, 0, s->cluster_size);
+        s->refcount_block_cache_offset = new_block;
     }
 
-    return refcount_block_offset;
+    /* Now the new refcount block needs to be written to disk */
+    ret = bdrv_pwrite_sync(s->hd, new_block, s->refcount_block_cache,
+        s->cluster_size);
+    if (ret < 0) {
+        goto fail_block;
+    }
+
+    /* If the refcount table is big enough, just hook the block up there */
+    if (refcount_table_index < s->refcount_table_size) {
+        uint64_t data64 = cpu_to_be64(new_block);
+        ret = bdrv_pwrite_sync(s->hd,
+            s->refcount_table_offset + refcount_table_index * sizeof(uint64_t),
+            &data64, sizeof(data64));
+        if (ret < 0) {
+            goto fail_block;
+        }
+
+        s->refcount_table[refcount_table_index] = new_block;
+        return new_block;
+    }
+
+    /*
+     * If we come here, we need to grow the refcount table. Again, a new
+     * refcount table needs some space and we can't simply allocate to avoid
+     * endless recursion.
+     *
+     * Therefore let's grab new refcount blocks at the end of the image, which
+     * will describe themselves and the new refcount table. This way we can
+     * reference them only in the new table and do the switch to the new
+     * refcount table at once without producing an inconsistent state in
+     * between.
+     */
+    /* Calculate the number of refcount blocks needed so far */
+    uint64_t refcount_block_clusters = 1 << (s->cluster_bits - REFCOUNT_SHIFT);
+    uint64_t blocks_used = (s->free_cluster_index +
+        refcount_block_clusters - 1) / refcount_block_clusters;
+
+    /* And now we need at least one block more for the new metadata */
+    uint64_t table_size = next_refcount_table_size(s, blocks_used + 1);
+    uint64_t last_table_size;
+    uint64_t blocks_clusters;
+    do {
+        uint64_t table_clusters = size_to_clusters(s, table_size);
+        blocks_clusters = 1 +
+            ((table_clusters + refcount_block_clusters - 1)
+            / refcount_block_clusters);
+        uint64_t meta_clusters = table_clusters + blocks_clusters;
+
+        last_table_size = table_size;
+        table_size = next_refcount_table_size(s, blocks_used +
+            ((meta_clusters + refcount_block_clusters - 1)
+            / refcount_block_clusters));
+
+    } while (last_table_size != table_size);
+
+#ifdef DEBUG_ALLOC2
+    fprintf(stderr, "qcow2: Grow refcount table %" PRId32 " => %" PRId64 "\n",
+        s->refcount_table_size, table_size);
+#endif
+
+    /* Create the new refcount table and blocks */
+    uint64_t meta_offset = (blocks_used * refcount_block_clusters) *
+        s->cluster_size;
+    uint64_t table_offset = meta_offset + blocks_clusters * s->cluster_size;
+    uint16_t *new_blocks = qemu_mallocz(blocks_clusters * s->cluster_size);
+    uint64_t *new_table = qemu_mallocz(table_size * sizeof(uint64_t));
+
+    assert(meta_offset >= (s->free_cluster_index * s->cluster_size));
+
+    /* Fill the new refcount table */
+    memcpy(new_table, s->refcount_table,
+        s->refcount_table_size * sizeof(uint64_t));
+    new_table[refcount_table_index] = new_block;
+
+    int i;
+    for (i = 0; i < blocks_clusters; i++) {
+        new_table[blocks_used + i] = meta_offset + (i * s->cluster_size);
+    }
+
+    /* Fill the refcount blocks */
+    uint64_t table_clusters = size_to_clusters(s, table_size * sizeof(uint64_t));
+    int block = 0;
+    for (i = 0; i < table_clusters + blocks_clusters; i++) {
+        new_blocks[block++] = cpu_to_be16(1);
+    }
+
+    /* Write refcount blocks to disk */
+    ret = bdrv_pwrite_sync(s->hd, meta_offset, new_blocks,
+        blocks_clusters * s->cluster_size);
+    qemu_free(new_blocks);
+    if (ret < 0) {
+        goto fail_table;
+    }
+
+    /* Write refcount table to disk */
+    for(i = 0; i < table_size; i++) {
+        cpu_to_be64s(&new_table[i]);
+    }
+
+    ret = bdrv_pwrite_sync(s->hd, table_offset, new_table,
+        table_size * sizeof(uint64_t));
+    if (ret < 0) {
+        goto fail_table;
+    }
+
+    for(i = 0; i < table_size; i++) {
+        cpu_to_be64s(&new_table[i]);
+    }
+
+    /* Hook up the new refcount table in the qcow2 header */
+    uint8_t data[12];
+    cpu_to_be64w((uint64_t*)data, table_offset);
+    cpu_to_be32w((uint32_t*)(data + 8), table_clusters);
+    ret = bdrv_pwrite_sync(s->hd, offsetof(QCowHeader, refcount_table_offset),
+        data, sizeof(data));
+    if (ret < 0) {
+        goto fail_table;
+    }
+
+    /* And switch it in memory */
+    uint64_t old_table_offset = s->refcount_table_offset;
+    uint64_t old_table_size = s->refcount_table_size;
+
+    qemu_free(s->refcount_table);
+    s->refcount_table = new_table;
+    s->refcount_table_size = table_size;
+    s->refcount_table_offset = table_offset;
+
+    /* Free old table. Remember, we must not change free_cluster_index */
+    uint64_t old_free_cluster_index = s->free_cluster_index;
+    qcow2_free_clusters(bs, old_table_offset, old_table_size * sizeof(uint64_t));
+    s->free_cluster_index = old_free_cluster_index;
+
+    ret = load_refcount_block(bs, new_block);
+    if (ret < 0) {
+        goto fail_block;
+    }
+
+    return new_block;
+
+fail_table:
+    qemu_free(new_table);
+fail_block:
+    s->refcount_block_cache_offset = 0;
+    return ret;
 }
 
 #define REFCOUNTS_PER_SECTOR (512 >> REFCOUNT_SHIFT)
@@ -245,43 +397,52 @@ static int write_refcount_block_entries(BDRVQcowState *s,
     int64_t refcount_block_offset, int first_index, int last_index)
 {
     size_t size;
+    int ret;
 
     if (cache_refcount_updates) {
         return 0;
     }
 
+    if (first_index < 0) {
+        return 0;
+    }
+
     first_index &= ~(REFCOUNTS_PER_SECTOR - 1);
     last_index = (last_index + REFCOUNTS_PER_SECTOR)
         & ~(REFCOUNTS_PER_SECTOR - 1);
 
     size = (last_index - first_index) << REFCOUNT_SHIFT;
-    if (bdrv_pwrite(s->hd,
+    ret = bdrv_pwrite_sync(s->hd,
         refcount_block_offset + (first_index << REFCOUNT_SHIFT),
-        &s->refcount_block_cache[first_index], size) != size)
+        &s->refcount_block_cache[first_index], size);
-    {
+    if (ret < 0) {
-        return -EIO;
+        return ret;
     }
 
     return 0;
 }
 
 /* XXX: cache several refcount block clusters ? */
-static int update_refcount(BlockDriverState *bs,
+static int QEMU_WARN_UNUSED_RESULT update_refcount(BlockDriverState *bs,
-                            int64_t offset, int64_t length,
+    int64_t offset, int64_t length, int addend)
-                            int addend)
 {
     BDRVQcowState *s = bs->opaque;
     int64_t start, last, cluster_offset;
     int64_t refcount_block_offset = 0;
     int64_t table_index = -1, old_table_index;
     int first_index = -1, last_index = -1;
+    int ret;
 
 #ifdef DEBUG_ALLOC2
     printf("update_refcount: offset=%" PRId64 " size=%" PRId64 " addend=%d\n",
            offset, length, addend);
 #endif
-    if (length <= 0)
+    if (length < 0) {
         return -EINVAL;
+    } else if (length == 0) {
+        return 0;
+    }
+
     start = offset & ~(s->cluster_size - 1);
     last = (offset + length - 1) & ~(s->cluster_size - 1);
     for(cluster_offset = start; cluster_offset <= last;
@@ -289,6 +450,7 @@ static int update_refcount(BlockDriverState *bs,
     {
         int block_index, refcount;
         int64_t cluster_index = cluster_offset >> s->cluster_bits;
+        int64_t new_block;
 
         /* Only write refcount block to disk when we are done with it */
         old_table_index = table_index;
@@ -306,10 +468,12 @@ static int update_refcount(BlockDriverState *bs,
         }
 
         /* Load the refcount block and allocate it if needed */
-        refcount_block_offset = alloc_refcount_block(bs, cluster_index);
+        new_block = alloc_refcount_block(bs, cluster_index);
-        if (refcount_block_offset < 0) {
+        if (new_block < 0) {
-            return refcount_block_offset;
+            ret = new_block;
+            goto fail;
         }
+        refcount_block_offset = new_block;
 
         /* we can update the count and save it */
         block_index = cluster_index &
@@ -323,24 +487,38 @@ static int update_refcount(BlockDriverState *bs,
 
         refcount = be16_to_cpu(s->refcount_block_cache[block_index]);
         refcount += addend;
-        if (refcount < 0 || refcount > 0xffff)
+        if (refcount < 0 || refcount > 0xffff) {
-            return -EINVAL;
+            ret = -EINVAL;
+            goto fail;
+        }
         if (refcount == 0 && cluster_index < s->free_cluster_index) {
             s->free_cluster_index = cluster_index;
         }
         s->refcount_block_cache[block_index] = cpu_to_be16(refcount);
     }
 
+    ret = 0;
+fail:
+
     /* Write last changed block to disk */
     if (refcount_block_offset != 0) {
         if (write_refcount_block_entries(s, refcount_block_offset,
             first_index, last_index) < 0)
         {
-            return -EIO;
+            return ret < 0 ? ret : -EIO;
         }
     }
 
-    return 0;
+    /*
+     * Try do undo any updates if an error is returned (This may succeed in
+     * some cases like ENOSPC for allocating a new refcount block)
+     */
+    if (ret < 0) {
+        int dummy;
+        dummy = update_refcount(bs, offset, cluster_offset - offset, -addend);
+    }
+
+    return ret;
 }
 
 /* addend must be 1 or -1 */
@@ -390,9 +568,13 @@ retry:
 int64_t qcow2_alloc_clusters(BlockDriverState *bs, int64_t size)
 {
     int64_t offset;
+    int ret;
 
     offset = alloc_clusters_noref(bs, size);
-    update_refcount(bs, offset, size, 1);
+    ret = update_refcount(bs, offset, size, 1);
+    if (ret < 0) {
+        return ret;
+    }
     return offset;
 }
 
@@ -407,6 +589,9 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
     assert(size > 0 && size <= s->cluster_size);
     if (s->free_byte_offset == 0) {
         s->free_byte_offset = qcow2_alloc_clusters(bs, s->cluster_size);
+        if (s->free_byte_offset < 0) {
+            return s->free_byte_offset;
+        }
     }
  redo:
     free_in_cluster = s->cluster_size -
@@ -422,6 +607,9 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
             update_cluster_refcount(bs, offset >> s->cluster_bits, 1);
     } else {
         offset = qcow2_alloc_clusters(bs, s->cluster_size);
+        if (offset < 0) {
+            return offset;
+        }
         cluster_offset = s->free_byte_offset & ~(s->cluster_size - 1);
         if ((cluster_offset + s->cluster_size) == offset) {
             /* we are lucky: contiguous data */
@@ -439,7 +627,13 @@ int64_t qcow2_alloc_bytes(BlockDriverState *bs, int size)
 void qcow2_free_clusters(BlockDriverState *bs,
                           int64_t offset, int64_t size)
 {
-    update_refcount(bs, offset, size, -1);
+    int ret;
+
+    ret = update_refcount(bs, offset, size, -1);
+    if (ret < 0) {
+        fprintf(stderr, "qcow2_free_clusters failed: %s\n", strerror(-ret));
+        /* TODO Remember the clusters to free them later and avoid leaking */
+    }
 }
 
 /*
@@ -549,9 +743,15 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
                     if (offset & QCOW_OFLAG_COMPRESSED) {
                         nb_csectors = ((offset >> s->csize_shift) &
                                        s->csize_mask) + 1;
-                        if (addend != 0)
+                        if (addend != 0) {
-                            update_refcount(bs, (offset & s->cluster_offset_mask) & ~511,
+                            int ret;
-                                            nb_csectors * 512, addend);
+                            ret = update_refcount(bs,
+                                (offset & s->cluster_offset_mask) & ~511,
+                                nb_csectors * 512, addend);
+                            if (ret < 0) {
+                                goto fail;
+                            }
+                        }
                         /* compressed clusters are never modified */
                         refcount = 2;
                     } else {
@@ -572,8 +772,8 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
                 }
             }
             if (l2_modified) {
-                if (bdrv_pwrite(s->hd,
+                if (bdrv_pwrite_sync(s->hd,
-                                l2_offset, l2_table, l2_size) != l2_size)
+                                l2_offset, l2_table, l2_size) < 0)
                     goto fail;
             }
 
@@ -594,8 +794,8 @@ int qcow2_update_snapshot_refcount(BlockDriverState *bs,
     if (l1_modified) {
         for(i = 0; i < l1_size; i++)
             cpu_to_be64s(&l1_table[i]);
-        if (bdrv_pwrite(s->hd, l1_table_offset, l1_table,
+        if (bdrv_pwrite_sync(s->hd, l1_table_offset, l1_table,
-                        l1_size2) != l1_size2)
+                        l1_size2) < 0)
             goto fail;
         for(i = 0; i < l1_size; i++)
             be64_to_cpus(&l1_table[i]);

block/qcow2-snapshot.c

@@ -139,6 +139,9 @@ static int qcow_write_snapshots(BlockDriverState *bs)
 
     snapshots_offset = qcow2_alloc_clusters(bs, snapshots_size);
     offset = snapshots_offset;
+    if (offset < 0) {
+        return offset;
+    }
 
     for(i = 0; i < s->nb_snapshots; i++) {
         sn = s->snapshots + i;
@@ -155,25 +158,25 @@ static int qcow_write_snapshots(BlockDriverState *bs)
         h.id_str_size = cpu_to_be16(id_str_size);
         h.name_size = cpu_to_be16(name_size);
         offset = align_offset(offset, 8);
-        if (bdrv_pwrite(s->hd, offset, &h, sizeof(h)) != sizeof(h))
+        if (bdrv_pwrite_sync(s->hd, offset, &h, sizeof(h)) < 0)
             goto fail;
         offset += sizeof(h);
-        if (bdrv_pwrite(s->hd, offset, sn->id_str, id_str_size) != id_str_size)
+        if (bdrv_pwrite_sync(s->hd, offset, sn->id_str, id_str_size) < 0)
             goto fail;
         offset += id_str_size;
-        if (bdrv_pwrite(s->hd, offset, sn->name, name_size) != name_size)
+        if (bdrv_pwrite_sync(s->hd, offset, sn->name, name_size) < 0)
             goto fail;
         offset += name_size;
     }
 
     /* update the various header fields */
     data64 = cpu_to_be64(snapshots_offset);
-    if (bdrv_pwrite(s->hd, offsetof(QCowHeader, snapshots_offset),
+    if (bdrv_pwrite_sync(s->hd, offsetof(QCowHeader, snapshots_offset),
-                    &data64, sizeof(data64)) != sizeof(data64))
+                    &data64, sizeof(data64)) < 0)
         goto fail;
     data32 = cpu_to_be32(s->nb_snapshots);
-    if (bdrv_pwrite(s->hd, offsetof(QCowHeader, nb_snapshots),
+    if (bdrv_pwrite_sync(s->hd, offsetof(QCowHeader, nb_snapshots),
-                    &data32, sizeof(data32)) != sizeof(data32))
+                    &data32, sizeof(data32)) < 0)
         goto fail;
 
     /* free the old snapshot table */
@@ -235,6 +238,7 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
     QCowSnapshot *snapshots1, sn1, *sn = &sn1;
     int i, ret;
     uint64_t *l1_table = NULL;
+    int64_t l1_table_offset;
 
     memset(sn, 0, sizeof(*sn));
 
@@ -263,7 +267,12 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
         goto fail;
 
     /* create the L1 table of the snapshot */
-    sn->l1_table_offset = qcow2_alloc_clusters(bs, s->l1_size * sizeof(uint64_t));
+    l1_table_offset = qcow2_alloc_clusters(bs, s->l1_size * sizeof(uint64_t));
+    if (l1_table_offset < 0) {
+        goto fail;
+    }
+
+    sn->l1_table_offset = l1_table_offset;
     sn->l1_size = s->l1_size;
 
     if (s->l1_size != 0) {
@@ -275,9 +284,8 @@ int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info)
     for(i = 0; i < s->l1_size; i++) {
         l1_table[i] = cpu_to_be64(s->l1_table[i]);
     }
-    if (bdrv_pwrite(s->hd, sn->l1_table_offset,
+    if (bdrv_pwrite_sync(s->hd, sn->l1_table_offset,
-                    l1_table, s->l1_size * sizeof(uint64_t)) !=
+                    l1_table, s->l1_size * sizeof(uint64_t)) < 0)
-        (s->l1_size * sizeof(uint64_t)))
         goto fail;
     qemu_free(l1_table);
     l1_table = NULL;
@@ -326,8 +334,8 @@ int qcow2_snapshot_goto(BlockDriverState *bs, const char *snapshot_id)
     if (bdrv_pread(s->hd, sn->l1_table_offset,
                    s->l1_table, l1_size2) != l1_size2)
         goto fail;
-    if (bdrv_pwrite(s->hd, s->l1_table_offset,
+    if (bdrv_pwrite_sync(s->hd, s->l1_table_offset,
-                    s->l1_table, l1_size2) != l1_size2)
+                    s->l1_table, l1_size2) < 0)
         goto fail;
     for(i = 0;i < s->l1_size; i++) {
         be64_to_cpus(&s->l1_table[i]);

block/qcow2.c

@@ -467,8 +467,10 @@ static void qcow_aio_read_cb(void *opaque, int ret)
         acb->hd_aiocb = bdrv_aio_readv(s->hd,
                             (acb->cluster_offset >> 9) + index_in_cluster,
                             &acb->hd_qiov, acb->n, qcow_aio_read_cb, acb);
-        if (acb->hd_aiocb == NULL)
+        if (acb->hd_aiocb == NULL) {
+            ret = -EIO;
             goto done;
+        }
     }
 
     return;
@@ -561,7 +563,7 @@ static void qcow_aio_write_cb(void *opaque, int ret)
     acb->hd_aiocb = NULL;
 
     if (ret >= 0) {
-        ret = qcow2_alloc_cluster_link_l2(bs, acb->cluster_offset, &acb->l2meta);
+        ret = qcow2_alloc_cluster_link_l2(bs, &acb->l2meta);
     }
 
     run_dependent_requests(&acb->l2meta);
@@ -585,21 +587,23 @@ static void qcow_aio_write_cb(void *opaque, int ret)
         n_end > QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors)
         n_end = QCOW_MAX_CRYPT_CLUSTERS * s->cluster_sectors;
 
-    acb->cluster_offset = qcow2_alloc_cluster_offset(bs, acb->sector_num << 9,
+    ret = qcow2_alloc_cluster_offset(bs, acb->sector_num << 9,
-                                          index_in_cluster,
+        index_in_cluster, n_end, &acb->n, &acb->l2meta);
-                                          n_end, &acb->n, &acb->l2meta);
+    if (ret < 0) {
+        goto done;
+    }
+
+    acb->cluster_offset = acb->l2meta.cluster_offset;
 
     /* Need to wait for another request? If so, we are done for now. */
-    if (!acb->cluster_offset && acb->l2meta.depends_on != NULL) {
+    if (acb->l2meta.nb_clusters == 0 && acb->l2meta.depends_on != NULL) {
         QLIST_INSERT_HEAD(&acb->l2meta.depends_on->dependent_requests,
             acb, next_depend);
         return;
     }
 
-    if (!acb->cluster_offset || (acb->cluster_offset & 511) != 0) {
+    assert((acb->cluster_offset & 511) == 0);
-        ret = -EIO;
+
-        goto done;
-    }
     if (s->crypt_method) {
         if (!acb->cluster_data) {
             acb->cluster_data = qemu_mallocz(QCOW_MAX_CRYPT_CLUSTERS *
@@ -618,11 +622,17 @@ static void qcow_aio_write_cb(void *opaque, int ret)
                                     (acb->cluster_offset >> 9) + index_in_cluster,
                                     &acb->hd_qiov, acb->n,
                                     qcow_aio_write_cb, acb);
-    if (acb->hd_aiocb == NULL)
+    if (acb->hd_aiocb == NULL) {
-        goto done;
+        ret = -EIO;
+        goto fail;
+    }
 
     return;
 
+fail:
+    if (acb->l2meta.nb_clusters != 0) {
+        QLIST_REMOVE(&acb->l2meta, next_in_flight);
+    }
 done:
     if (acb->qiov->niov > 1)
         qemu_vfree(acb->orig_buf);
@@ -683,27 +693,27 @@ static int get_bits_from_size(size_t size)
 static int preallocate(BlockDriverState *bs)
 {
     BDRVQcowState *s = bs->opaque;
-    uint64_t cluster_offset = 0;
     uint64_t nb_sectors;
     uint64_t offset;
     int num;
+    int ret;
     QCowL2Meta meta;
 
     nb_sectors = bdrv_getlength(bs) >> 9;
     offset = 0;
     QLIST_INIT(&meta.dependent_requests);
+    meta.cluster_offset = 0;
 
     while (nb_sectors) {
         num = MIN(nb_sectors, INT_MAX >> 9);
-        cluster_offset = qcow2_alloc_cluster_offset(bs, offset, 0, num, &num,
+        ret = qcow2_alloc_cluster_offset(bs, offset, 0, num, &num, &meta);
-            &meta);
 
-        if (cluster_offset == 0) {
+        if (ret < 0) {
             return -1;
         }
 
-        if (qcow2_alloc_cluster_link_l2(bs, cluster_offset, &meta) < 0) {
+        if (qcow2_alloc_cluster_link_l2(bs, &meta) < 0) {
-            qcow2_free_any_clusters(bs, cluster_offset, meta.nb_clusters);
+            qcow2_free_any_clusters(bs, meta.cluster_offset, meta.nb_clusters);
             return -1;
         }
 
@@ -722,10 +732,10 @@ static int preallocate(BlockDriverState *bs)
      * all of the allocated clusters (otherwise we get failing reads after
      * EOF). Extend the image to the last allocated sector.
      */
-    if (cluster_offset != 0) {
+    if (meta.cluster_offset != 0) {
         uint8_t buf[512];
         memset(buf, 0, 512);
-        bdrv_write(s->hd, (cluster_offset >> 9) + num - 1, buf, 1);
+        bdrv_write(s->hd, (meta.cluster_offset >> 9) + num - 1, buf, 1);
     }
 
     return 0;
@@ -737,10 +747,11 @@ static int qcow_create2(const char *filename, int64_t total_size,
 {
 
     int fd, header_size, backing_filename_len, l1_size, i, shift, l2_bits;
-    int ref_clusters, backing_format_len = 0;
+    int ref_clusters, reftable_clusters, backing_format_len = 0;
     int rounded_ext_bf_len = 0;
     QCowHeader header;
     uint64_t tmp, offset;
+    uint64_t old_ref_clusters;
     QCowCreateState s1, *s = &s1;
     QCowExtension ext_bf = {0, 0};
 
@@ -799,17 +810,37 @@ static int qcow_create2(const char *filename, int64_t total_size,
     header.l1_size = cpu_to_be32(l1_size);
     offset += align_offset(l1_size * sizeof(uint64_t), s->cluster_size);
 
-    s->refcount_table = qemu_mallocz(s->cluster_size);
+    /* count how many refcount blocks needed */
+
+#define NUM_CLUSTERS(bytes) \
+    (((bytes) + (s->cluster_size) - 1) / (s->cluster_size))
+
+    ref_clusters = NUM_CLUSTERS(NUM_CLUSTERS(offset) * sizeof(uint16_t));
+
+    do {
+        uint64_t image_clusters;
+        old_ref_clusters = ref_clusters;
+
+        /* Number of clusters used for the refcount table */
+        reftable_clusters = NUM_CLUSTERS(ref_clusters * sizeof(uint64_t));
+
+        /* Number of clusters that the whole image will have */
+        image_clusters = NUM_CLUSTERS(offset) + ref_clusters
+            + reftable_clusters;
+
+        /* Number of refcount blocks needed for the image */
+        ref_clusters = NUM_CLUSTERS(image_clusters * sizeof(uint16_t));
+
+    } while (ref_clusters != old_ref_clusters);
+
+    s->refcount_table = qemu_mallocz(reftable_clusters * s->cluster_size);
 
     s->refcount_table_offset = offset;
     header.refcount_table_offset = cpu_to_be64(offset);
-    header.refcount_table_clusters = cpu_to_be32(1);
+    header.refcount_table_clusters = cpu_to_be32(reftable_clusters);
-    offset += s->cluster_size;
+    offset += (reftable_clusters * s->cluster_size);
     s->refcount_block_offset = offset;
 
-    /* count how many refcount blocks needed */
-    tmp = offset >> s->cluster_bits;
-    ref_clusters = (tmp >> (s->cluster_bits - REFCOUNT_SHIFT)) + 1;
     for (i=0; i < ref_clusters; i++) {
         s->refcount_table[i] = cpu_to_be64(offset);
         offset += s->cluster_size;
@@ -821,7 +852,8 @@ static int qcow_create2(const char *filename, int64_t total_size,
     qcow2_create_refcount_update(s, 0, header_size);
     qcow2_create_refcount_update(s, s->l1_table_offset,
         l1_size * sizeof(uint64_t));
-    qcow2_create_refcount_update(s, s->refcount_table_offset, s->cluster_size);
+    qcow2_create_refcount_update(s, s->refcount_table_offset,
+        reftable_clusters * s->cluster_size);
     qcow2_create_refcount_update(s, s->refcount_block_offset,
         ref_clusters * s->cluster_size);
 
@@ -849,7 +881,8 @@ static int qcow_create2(const char *filename, int64_t total_size,
         write(fd, &tmp, sizeof(tmp));
     }
     lseek(fd, s->refcount_table_offset, SEEK_SET);
-    write(fd, s->refcount_table, s->cluster_size);
+    write(fd, s->refcount_table,
+        reftable_clusters * s->cluster_size);
 
     lseek(fd, s->refcount_block_offset, SEEK_SET);
     write(fd, s->refcount_block, ref_clusters * s->cluster_size);
@@ -1056,12 +1089,13 @@ static int qcow_save_vmstate(BlockDriverState *bs, const uint8_t *buf,
 {
     BDRVQcowState *s = bs->opaque;
     int growable = bs->growable;
+    int ret;
 
     bs->growable = 1;
-    bdrv_pwrite(bs, qcow_vm_state_offset(s) + pos, buf, size);
+    ret = bdrv_pwrite(bs, qcow_vm_state_offset(s) + pos, buf, size);
     bs->growable = growable;
 
-    return size;
+    return ret;
 }
 
 static int qcow_load_vmstate(BlockDriverState *bs, uint8_t *buf,

block/qcow2.h

@@ -135,6 +135,7 @@ struct QCowAIOCB;
 typedef struct QCowL2Meta
 {
     uint64_t offset;
+    uint64_t cluster_offset;
     int n_start;
     int nb_available;
     int nb_clusters;
@@ -191,16 +192,13 @@ void qcow2_encrypt_sectors(BDRVQcowState *s, int64_t sector_num,
 
 uint64_t qcow2_get_cluster_offset(BlockDriverState *bs, uint64_t offset,
     int *num);
-uint64_t qcow2_alloc_cluster_offset(BlockDriverState *bs,
+int qcow2_alloc_cluster_offset(BlockDriverState *bs, uint64_t offset,
-                              uint64_t offset,
+    int n_start, int n_end, int *num, QCowL2Meta *m);
-                              int n_start, int n_end,
-                              int *num, QCowL2Meta *m);
 uint64_t qcow2_alloc_compressed_cluster_offset(BlockDriverState *bs,
                                          uint64_t offset,
                                          int compressed_size);
 
-int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, uint64_t cluster_offset,
+int qcow2_alloc_cluster_link_l2(BlockDriverState *bs, QCowL2Meta *m);
-    QCowL2Meta *m);
 
 /* qcow2-snapshot.c functions */
 int qcow2_snapshot_create(BlockDriverState *bs, QEMUSnapshotInfo *sn_info);

block/raw-posix.c

@@ -27,6 +27,8 @@
 #include "qemu-log.h"
 #include "block_int.h"
 #include "module.h"
+#include "compatfd.h"
+#include <assert.h>
 #include "block/raw-posix-aio.h"
 
 #ifdef CONFIG_COCOA
@@ -595,7 +597,7 @@ static void raw_close(BlockDriverState *bs)
         close(s->fd);
         s->fd = -1;
         if (s->aligned_buf != NULL)
-            qemu_free(s->aligned_buf);
+            qemu_vfree(s->aligned_buf);
     }
 }
 

block/vdi.c

@@ -399,6 +399,15 @@ static int vdi_open(BlockDriverState *bs, const char *filename, int flags)
     vdi_header_print(&header);
 #endif
 
+    if (header.disk_size % SECTOR_SIZE != 0) {
+        /* 'VBoxManage convertfromraw' can create images with odd disk sizes.
+           We accept them but round the disk size to the next multiple of
+           SECTOR_SIZE. */
+        logout("odd disk size %" PRIu64 " B, round up\n", header.disk_size);
+        header.disk_size += SECTOR_SIZE - 1;
+        header.disk_size &= ~(SECTOR_SIZE - 1);
+    }
+
     if (header.version != VDI_VERSION_1_1) {
         logout("unsupported version %u.%u\n",
                header.version >> 16, header.version & 0xffff);
@@ -417,9 +426,9 @@ static int vdi_open(BlockDriverState *bs, const char *filename, int flags)
     } else if (header.block_size != 1 * MiB) {
         logout("unsupported block size %u B\n", header.block_size);
         goto fail;
-    } else if (header.disk_size !=
+    } else if (header.disk_size >
                (uint64_t)header.blocks_in_image * header.block_size) {
-        logout("unexpected block number %u B\n", header.blocks_in_image);
+        logout("unsupported disk size %" PRIu64 " B\n", header.disk_size);
         goto fail;
     } else if (!uuid_is_null(header.uuid_link)) {
         logout("link uuid != 0, unsupported\n");
@@ -831,7 +840,10 @@ static int vdi_create(const char *filename, QEMUOptionParameter *options)
         return -errno;
     }
 
-    blocks = bytes / block_size;
+    /* We need enough blocks to store the given disk size,
+       so always round up. */
+    blocks = (bytes + block_size - 1) / block_size;
+
     bmap_size = blocks * sizeof(uint32_t);
     bmap_size = ((bmap_size + SECTOR_SIZE - 1) & ~(SECTOR_SIZE -1));
 

block/vmdk.c

@@ -87,14 +87,6 @@ typedef struct VmdkMetaData {
     int valid;
 } VmdkMetaData;
 
-typedef struct ActiveBDRVState{
-    BlockDriverState *hd;            // active image handler
-    uint64_t cluster_offset;         // current write offset
-}ActiveBDRVState;
-
-static ActiveBDRVState activeBDRV;
-
-
 static int vmdk_probe(const uint8_t *buf, int buf_size, const char *filename)
 {
     uint32_t magic;
@@ -161,7 +153,7 @@ static int vmdk_write_cid(BlockDriverState *bs, uint32_t cid)
         pstrcat(desc, sizeof(desc), tmp_desc);
     }
 
-    if (bdrv_pwrite(s->hd, 0x200, desc, DESC_SIZE) != DESC_SIZE)
+    if (bdrv_pwrite_sync(s->hd, 0x200, desc, DESC_SIZE) < 0)
         return -1;
     return 0;
 }
@@ -285,7 +277,6 @@ static int vmdk_snapshot_create(const char *filename, const char *backing_file)
         goto fail_rgd;
     if (write(snp_fd, rgd_buf, gd_size) == -1)
         goto fail_rgd;
-    qemu_free(rgd_buf);
 
     /* write GD */
     gd_buf = qemu_malloc(gd_size);
@@ -298,6 +289,7 @@ static int vmdk_snapshot_create(const char *filename, const char *backing_file)
     if (write(snp_fd, gd_buf, gd_size) == -1)
         goto fail_gd;
     qemu_free(gd_buf);
+    qemu_free(rgd_buf);
 
     close(p_fd);
     close(snp_fd);
@@ -458,30 +450,28 @@ static uint64_t get_cluster_offset(BlockDriverState *bs, VmdkMetaData *m_data,
 static int get_whole_cluster(BlockDriverState *bs, uint64_t cluster_offset,
                              uint64_t offset, int allocate)
 {
-    uint64_t parent_cluster_offset;
     BDRVVmdkState *s = bs->opaque;
     uint8_t  whole_grain[s->cluster_sectors*512];        // 128 sectors * 512 bytes each = grain size 64KB
 
     // we will be here if it's first write on non-exist grain(cluster).
     // try to read from parent image, if exist
     if (bs->backing_hd) {
-        BDRVVmdkState *ps = bs->backing_hd->opaque;
+        int ret;
 
         if (!vmdk_is_cid_valid(bs))
             return -1;
 
-        parent_cluster_offset = get_cluster_offset(bs->backing_hd, NULL,
+        ret = bdrv_read(bs->backing_hd, offset >> 9, whole_grain,
-            offset, allocate);
+            s->cluster_sectors);
+        if (ret < 0) {
+            return -1;
+        }
 
-        if (parent_cluster_offset) {
+        //Write grain only into the active image
-            BDRVVmdkState *act_s = activeBDRV.hd->opaque;
+        ret = bdrv_write(s->hd, cluster_offset, whole_grain,
-
+            s->cluster_sectors);
-            if (bdrv_pread(ps->hd, parent_cluster_offset, whole_grain, ps->cluster_sectors*512) != ps->cluster_sectors*512)
+        if (ret < 0) {
-                return -1;
+            return -1;
-
-            //Write grain only into the active image
-            if (bdrv_pwrite(act_s->hd, activeBDRV.cluster_offset << 9, whole_grain, sizeof(whole_grain)) != sizeof(whole_grain))
-                return -1;
         }
     }
     return 0;
@@ -492,14 +482,14 @@ static int vmdk_L2update(BlockDriverState *bs, VmdkMetaData *m_data)
     BDRVVmdkState *s = bs->opaque;
 
     /* update L2 table */
-    if (bdrv_pwrite(s->hd, ((int64_t)m_data->l2_offset * 512) + (m_data->l2_index * sizeof(m_data->offset)),
+    if (bdrv_pwrite_sync(s->hd, ((int64_t)m_data->l2_offset * 512) + (m_data->l2_index * sizeof(m_data->offset)),
-                    &(m_data->offset), sizeof(m_data->offset)) != sizeof(m_data->offset))
+                    &(m_data->offset), sizeof(m_data->offset)) < 0)
         return -1;
     /* update backup L2 table */
     if (s->l1_backup_table_offset != 0) {
         m_data->l2_offset = s->l1_backup_table[m_data->l1_index];
-        if (bdrv_pwrite(s->hd, ((int64_t)m_data->l2_offset * 512) + (m_data->l2_index * sizeof(m_data->offset)),
+        if (bdrv_pwrite_sync(s->hd, ((int64_t)m_data->l2_offset * 512) + (m_data->l2_index * sizeof(m_data->offset)),
-                        &(m_data->offset), sizeof(m_data->offset)) != sizeof(m_data->offset))
+                        &(m_data->offset), sizeof(m_data->offset)) < 0)
             return -1;
     }
 
@@ -567,9 +557,6 @@ static uint64_t get_cluster_offset(BlockDriverState *bs, VmdkMetaData *m_data,
             cluster_offset >>= 9;
             tmp = cpu_to_le32(cluster_offset);
             l2_table[l2_index] = tmp;
-            // Save the active image state
-            activeBDRV.cluster_offset = cluster_offset;
-            activeBDRV.hd = bs;
         }
         /* First of all we write grain itself, to avoid race condition
          * that may to corrupt the image.

block/vpc.c

@@ -266,7 +266,7 @@ static inline int64_t get_sector_offset(BlockDriverState *bs,
 
         s->last_bitmap_offset = bitmap_offset;
         memset(bitmap, 0xff, s->bitmap_size);
-        bdrv_pwrite(s->hd, bitmap_offset, bitmap, s->bitmap_size);
+        bdrv_pwrite_sync(s->hd, bitmap_offset, bitmap, s->bitmap_size);
     }
 
 //    printf("sector: %" PRIx64 ", index: %x, offset: %x, bioff: %" PRIx64 ", bloff: %" PRIx64 "\n",
@@ -316,7 +316,7 @@ static int rewrite_footer(BlockDriverState* bs)
     BDRVVPCState *s = bs->opaque;
     int64_t offset = s->free_data_block_offset;
 
-    ret = bdrv_pwrite(s->hd, offset, s->footer_buf, HEADER_SIZE);
+    ret = bdrv_pwrite_sync(s->hd, offset, s->footer_buf, HEADER_SIZE);
     if (ret < 0)
         return ret;
 
@@ -351,7 +351,8 @@ static int64_t alloc_block(BlockDriverState* bs, int64_t sector_num)
 
     // Initialize the block's bitmap
     memset(bitmap, 0xff, s->bitmap_size);
-    bdrv_pwrite(s->hd, s->free_data_block_offset, bitmap, s->bitmap_size);
+    bdrv_pwrite_sync(s->hd, s->free_data_block_offset, bitmap,
+        s->bitmap_size);
 
     // Write new footer (the old one will be overwritten)
     s->free_data_block_offset += s->block_size + s->bitmap_size;
@@ -362,7 +363,7 @@ static int64_t alloc_block(BlockDriverState* bs, int64_t sector_num)
     // Write BAT entry to disk
     bat_offset = s->bat_offset + (4 * index);
     bat_value = be32_to_cpu(s->pagetable[index]);
-    ret = bdrv_pwrite(s->hd, bat_offset, &bat_value, 4);
+    ret = bdrv_pwrite_sync(s->hd, bat_offset, &bat_value, 4);
     if (ret < 0)
         goto fail;
 
@@ -470,9 +471,7 @@ static int calculate_geometry(int64_t total_sectors, uint16_t* cyls,
         }
     }
 
-    // Note: Rounding up deviates from the Virtual PC behaviour
+    *cyls = cyls_times_heads / *heads;
-    // However, we need this to avoid truncating images in qemu-img convert
-    *cyls = (cyls_times_heads + *heads - 1) / *heads;
 
     return 0;
 }
@@ -484,9 +483,9 @@ static int vpc_create(const char *filename, QEMUOptionParameter *options)
     struct vhd_dyndisk_header* dyndisk_header =
         (struct vhd_dyndisk_header*) buf;
     int fd, i;
-    uint16_t cyls;
+    uint16_t cyls = 0;
-    uint8_t heads;
+    uint8_t heads = 0;
-    uint8_t secs_per_cyl;
+    uint8_t secs_per_cyl = 0;
     size_t block_size, num_bat_entries;
     int64_t total_sectors = 0;
 
@@ -503,9 +502,14 @@ static int vpc_create(const char *filename, QEMUOptionParameter *options)
     if (fd < 0)
         return -EIO;
 
-    // Calculate matching total_size and geometry
+    /* Calculate matching total_size and geometry. Increase the number of
-    if (calculate_geometry(total_sectors, &cyls, &heads, &secs_per_cyl))
+       sectors requested until we get enough (or fail). */
-        return -EFBIG;
+    for (i = 0; total_sectors > (int64_t)cyls * heads * secs_per_cyl; i++) {
+        if (calculate_geometry(total_sectors + i,
+                               &cyls, &heads, &secs_per_cyl)) {
+            return -EFBIG;
+        }
+    }
     total_sectors = (int64_t) cyls * heads * secs_per_cyl;
 
     // Prepare the Hard Disk Footer

block/vvfat.c

@@ -868,7 +868,8 @@ static int init_directories(BDRVVVFATState* s,
     {
 	direntry_t* entry=array_get_next(&(s->directory));
 	entry->attributes=0x28; /* archive | volume label */
-	snprintf((char*)entry->name,11,"QEMU VVFAT");
+	memcpy(entry->name,"QEMU VVF",8);
+	memcpy(entry->extension,"AT ",3);
     }
 
     /* Now build FAT, and write back information into directory */
@@ -882,7 +883,7 @@ static int init_directories(BDRVVVFATState* s,
     mapping->dir_index = 0;
     mapping->info.dir.parent_mapping_index = -1;
     mapping->first_mapping_index = -1;
-    mapping->path = strdup(dirname);
+    mapping->path = qemu_strdup(dirname);
     i = strlen(mapping->path);
     if (i > 0 && mapping->path[i - 1] == '/')
 	mapping->path[i - 1] = '\0';
@@ -1632,10 +1633,10 @@ static uint32_t get_cluster_count_for_direntry(BDRVVVFATState* s,
 
 	    /* rename */
 	    if (strcmp(basename, basename2))
-		schedule_rename(s, cluster_num, strdup(path));
+		schedule_rename(s, cluster_num, qemu_strdup(path));
 	} else if (is_file(direntry))
 	    /* new file */
-	    schedule_new_file(s, strdup(path), cluster_num);
+	    schedule_new_file(s, qemu_strdup(path), cluster_num);
 	else {
 	    assert(0);
 	    return 0;
@@ -1752,10 +1753,10 @@ static int check_directory_consistency(BDRVVVFATState *s,
 	mapping->mode &= ~MODE_DELETED;
 
 	if (strcmp(basename, basename2))
-	    schedule_rename(s, cluster_num, strdup(path));
+	    schedule_rename(s, cluster_num, qemu_strdup(path));
     } else
 	/* new directory */
-	schedule_mkdir(s, cluster_num, strdup(path));
+	schedule_mkdir(s, cluster_num, qemu_strdup(path));
 
     lfn_init(&lfn);
     do {
@@ -2256,7 +2257,11 @@ static int commit_one_file(BDRVVVFATState* s,
 	c = c1;
     }
 
-    ftruncate(fd, size);
+    if (ftruncate(fd, size)) {
+        perror("ftruncate()");
+        close(fd);
+        return -4;
+    }
     close(fd);
 
     return commit_mappings(s, first_cluster, dir_index);

block_int.h

@@ -127,6 +127,7 @@ struct BlockDriverState {
     int64_t total_sectors; /* if we are reading a disk image, give its
                               size in sectors */
     int read_only; /* if true, the media is read only */
+    int open_flags; /* flags used to open the file, re-used for re-open */
     int removable; /* if true, the media can be removed */
     int locked;    /* if true, the media cannot temporarily be ejected */
     int encrypted; /* if true, the media is encrypted */

cache-utils.h

@@ -34,7 +34,28 @@ static inline void flush_icache_range(unsigned long start, unsigned long stop)
     asm volatile ("isync" : : : "memory");
 }
 
+/*
+ * Is this correct for PPC?
+ */
+static inline void dma_flush_range(unsigned long start, unsigned long stop)
+{
+}
+
+#elif defined(__ia64__)
+static inline void flush_icache_range(unsigned long start, unsigned long stop)
+{
+    while (start < stop) {
+	asm volatile ("fc %0" :: "r"(start));
+	start += 32;
+    }
+    asm volatile (";;sync.i;;srlz.i;;");
+}
+#define dma_flush_range(start, end) flush_icache_range(start, end)
+#define qemu_cache_utils_init(envp) do { (void) (envp); } while (0)
 #else
+static inline void dma_flush_range(unsigned long start, unsigned long stop)
+{
+}
 #define qemu_cache_utils_init(envp) do { (void) (envp); } while (0)
 #endif
 

check-qdict.c

@@ -205,6 +205,8 @@ START_TEST(qdict_put_exists_test)
 
     value = qdict_get_int(tests_dict, key);
     fail_unless(value == 2);
+
+    fail_unless(qdict_size(tests_dict) == 1);
 }
 END_TEST
 

compat/sys/eventfd.h

@@ -0,0 +1,13 @@
+#ifndef _COMPAT_SYS_EVENTFD
+#define _COMPAT_SYS_EVENTFD
+
+#include <unistd.h>
+#include <syscall.h>
+
+
+static inline int eventfd (int count, int flags)
+{
+    return syscall(SYS_eventfd, count, flags);
+}
+
+#endif

compatfd.c

@@ -0,0 +1,143 @@
+/*
+ * signalfd/eventfd compatibility
+ *
+ * Copyright IBM, Corp. 2008
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "qemu-common.h"
+#include "compatfd.h"
+
+#include <sys/syscall.h>
+#include <pthread.h>
+
+struct sigfd_compat_info
+{
+    sigset_t mask;
+    int fd;
+};
+
+static void *sigwait_compat(void *opaque)
+{
+    struct sigfd_compat_info *info = opaque;
+    int err;
+    sigset_t all;
+
+    sigfillset(&all);
+    sigprocmask(SIG_BLOCK, &all, NULL);
+
+    do {
+        siginfo_t siginfo;
+
+        err = sigwaitinfo(&info->mask, &siginfo);
+        if (err == -1 && errno == EINTR) {
+            err = 0;
+            continue;
+        }
+
+        if (err > 0) {
+            char buffer[128];
+            size_t offset = 0;
+
+            memcpy(buffer, &err, sizeof(err));
+            while (offset < sizeof(buffer)) {
+                ssize_t len;
+
+                len = write(info->fd, buffer + offset,
+                            sizeof(buffer) - offset);
+                if (len == -1 && errno == EINTR)
+                    continue;
+
+                if (len <= 0) {
+                    err = -1;
+                    break;
+                }
+
+                offset += len;
+            }
+        }
+    } while (err >= 0);
+
+    return NULL;
+}
+
+static int qemu_signalfd_compat(const sigset_t *mask)
+{
+    pthread_attr_t attr;
+    pthread_t tid;
+    struct sigfd_compat_info *info;
+    int fds[2];
+
+    info = malloc(sizeof(*info));
+    if (info == NULL) {
+        errno = ENOMEM;
+        return -1;
+    }
+
+    if (pipe(fds) == -1) {
+        free(info);
+        return -1;
+    }
+
+    qemu_set_cloexec(fds[0]);
+    qemu_set_cloexec(fds[1]);
+
+    memcpy(&info->mask, mask, sizeof(*mask));
+    info->fd = fds[1];
+
+    pthread_attr_init(&attr);
+    pthread_attr_setdetachstate(&attr, PTHREAD_CREATE_DETACHED);
+
+    pthread_create(&tid, &attr, sigwait_compat, info);
+
+    pthread_attr_destroy(&attr);
+
+    return fds[0];
+}
+
+int qemu_signalfd(const sigset_t *mask)
+{
+#if defined(CONFIG_SIGNALFD)
+    int ret;
+
+    ret = syscall(SYS_signalfd, -1, mask, _NSIG / 8);
+    if (ret != -1) {
+        qemu_set_cloexec(ret);
+        return ret;
+    }
+#endif
+
+    return qemu_signalfd_compat(mask);
+}
+
+int qemu_eventfd(int *fds)
+{
+    int ret;
+
+#if defined(CONFIG_EVENTFD)
+    ret = syscall(SYS_eventfd, 0);
+    if (ret >= 0) {
+        fds[0] = ret;
+        qemu_set_cloexec(ret);
+        if ((fds[1] = dup(ret)) == -1) {
+            close(ret);
+            return -1;
+        }
+        qemu_set_cloexec(fds[1]);
+        return 0;
+    }
+#endif
+
+    ret = pipe(fds);
+    if (ret != -1) {
+        qemu_set_cloexec(fds[0]);
+        qemu_set_cloexec(fds[1]);
+    }
+    return ret;
+}

compatfd.h

@@ -0,0 +1,45 @@
+/*
+ * signalfd/eventfd compatibility
+ *
+ * Copyright IBM, Corp. 2008
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#ifndef QEMU_COMPATFD_H
+#define QEMU_COMPATFD_H
+
+#include <signal.h>
+
+struct qemu_signalfd_siginfo {
+    uint32_t ssi_signo;   /* Signal number */
+    int32_t  ssi_errno;   /* Error number (unused) */
+    int32_t  ssi_code;    /* Signal code */
+    uint32_t ssi_pid;     /* PID of sender */
+    uint32_t ssi_uid;     /* Real UID of sender */
+    int32_t  ssi_fd;      /* File descriptor (SIGIO) */
+    uint32_t ssi_tid;     /* Kernel timer ID (POSIX timers) */
+    uint32_t ssi_band;    /* Band event (SIGIO) */
+    uint32_t ssi_overrun; /* POSIX timer overrun count */
+    uint32_t ssi_trapno;  /* Trap number that caused signal */
+    int32_t  ssi_status;  /* Exit status or signal (SIGCHLD) */
+    int32_t  ssi_int;     /* Integer sent by sigqueue(2) */
+    uint64_t ssi_ptr;     /* Pointer sent by sigqueue(2) */
+    uint64_t ssi_utime;   /* User CPU time consumed (SIGCHLD) */
+    uint64_t ssi_stime;   /* System CPU time consumed (SIGCHLD) */
+    uint64_t ssi_addr;    /* Address that generated signal
+                             (for hardware-generated signals) */
+    uint8_t  pad[48];     /* Pad size to 128 bytes (allow for
+                             additional fields in the future) */
+};
+
+int qemu_signalfd(const sigset_t *mask);
+
+int qemu_eventfd(int *fds);
+
+#endif

configure

@@ -160,7 +160,7 @@ else
   cpu=`uname -m`
 fi
 
-target_list=""
+target_list="x86_64-softmmu"
 case "$cpu" in
   alpha|cris|ia64|m68k|microblaze|ppc|ppc64|sparc64)
     cpu="$cpu"
@@ -197,6 +197,16 @@ case "$cpu" in
   ;;
 esac
 
+kvm_version() {
+    local fname="$(dirname "$0")/KVM_VERSION"
+
+    if test -f "$fname"; then
+        cat "$fname"
+    else
+        echo "qemu-kvm-devel"
+    fi
+}
+
 # Default value for a variable defining feature "foo".
 #  * foo="no"  feature will only be used if --enable-foo arg is given
 #  * foo=""    feature will be searched for, and if found, will be used
@@ -250,12 +260,18 @@ guest_base=""
 uname_release=""
 io_thread="no"
 mixemu="no"
+kvm_trace="no"
+kvm_cap_pit=""
+kvm_cap_device_assignment=""
 kerneldir=""
 aix="no"
 blobs="yes"
-pkgversion=""
+pkgversion=" ($(kvm_version))"
+cpu_emulation="yes"
+kvm_kmod="no"
 check_utests="no"
 user_pie="no"
+zero_malloc=""
 
 # OS specific
 if check_define __linux__ ; then
@@ -388,6 +404,13 @@ AIX)
   if [ "$cpu" = "i386" -o "$cpu" = "x86_64" ] ; then
     audio_possible_drivers="$audio_possible_drivers fmod"
   fi
+  if [ "$cpu" = "ia64" ] ; then
+     xen="no"
+     target_list="ia64-softmmu"
+     cpu_emulation="no"
+     gdbstub="no"
+     slirp="no"
+  fi
 ;;
 esac
 
@@ -517,6 +540,14 @@ for opt do
   ;;
   --enable-kvm) kvm="yes"
   ;;
+  --disable-kvm-cap-pit) kvm_cap_pit="no"
+  ;;
+  --enable-kvm-cap-pit) kvm_cap_pit="yes"
+  ;;
+  --disable-kvm-cap-device-assignment) kvm_cap_device_assignment="no"
+  ;;
+  --enable-kvm-cap-device-assignment) kvm_cap_device_assignment="yes"
+  ;;
   --enable-profiler) profiler="yes"
   ;;
   --enable-cocoa)
@@ -594,12 +625,16 @@ for opt do
   ;;
   --kerneldir=*) kerneldir="$optarg"
   ;;
+  --with-kvm-trace) kvm_trace="yes"
+  ;;
   --with-pkgversion=*) pkgversion=" ($optarg)"
   ;;
   --disable-docs) docs="no"
   ;;
   --enable-docs) docs="yes"
   ;;
+  --disable-cpu-emulation) cpu_emulation="no"
+  ;;
   *) echo "ERROR: unknown option $opt"; show_help="yes"
   ;;
   esac
@@ -721,6 +756,10 @@ echo "  --disable-bluez          disable bluez stack connectivity"
 echo "  --enable-bluez           enable bluez stack connectivity"
 echo "  --disable-kvm            disable KVM acceleration support"
 echo "  --enable-kvm             enable KVM acceleration support"
+echo "  --disable-cap-kvm-pit    disable KVM pit support"
+echo "  --enable-cap-kvm-pit     enable KVM pit support"
+echo "  --disable-cap-device-assignment    disable KVM device assignment support"
+echo "  --enable-cap-device-assignment     enable KVM device assignment support"
 echo "  --disable-nptl           disable usermode NPTL support"
 echo "  --enable-nptl            enable usermode NPTL support"
 echo "  --enable-system          enable all system emulation targets"
@@ -752,6 +791,8 @@ echo "  --enable-linux-aio       enable Linux AIO support"
 echo "  --enable-io-thread       enable IO thread"
 echo "  --disable-blobs          disable installing provided firmware blobs"
 echo "  --kerneldir=PATH         look for kernel includes in PATH"
+echo "  --with-kvm-trace         enable building the KVM module with the kvm trace option"
+echo "  --disable-cpu-emulation  disables use of qemu cpu emulation code"
 echo ""
 echo "NOTE: The object files are built at the place where configure is launched"
 exit 1
@@ -1383,8 +1424,22 @@ EOF
             kvm_cflags="$kvm_cflags -I$kerneldir/arch/$cpu/include"
       fi
   else
-      kvm_cflags=""
+      case "$cpu" in
+      i386 | x86_64)
+        kvm_arch="x86"
+        ;;
+      ppc)
+        kvm_arch="powerpc"
+        ;;
+      *)
+        kvm_arch="$cpu"
+        ;;
+      esac
+      kvm_cflags="-I$source_path/kvm/include"
+      kvm_cflags="$kvm_cflags -include $source_path/kvm/include/linux/config.h"
+      kvm_cflags="$kvm_cflags -I$source_path/kvm/include/$kvm_arch"
   fi
+  kvm_cflags="$kvm_cflags -idirafter $source_path/compat"
   if compile_prog "$kvm_cflags" "" ; then
     kvm=yes
   else
@@ -1406,6 +1461,75 @@ EOF
   fi
 fi
 
+##########################################
+# test for KVM_CAP_PIT
+
+if test "$kvm_cap_pit" != "no" ; then
+  if test "$kvm" = "no" -a "$kvm_cap_pit" = "yes" ; then
+      feature_not_found "kvm_cap_pit (kvm is not enabled)"
+  fi
+  cat > $TMPC <<EOF
+#include <linux/kvm.h>
+#ifndef KVM_CAP_PIT
+#error "kvm no pit capability"
+#endif
+int main(void) { return 0; }
+EOF
+  if compile_prog "$kvm_cflags" ""; then
+    kvm_cap_pit=yes
+  else
+    if test "$kvm_cap_pit" = "yes" ; then
+      feature_not_found "kvm_cap_pit"
+    fi
+    kvm_cap_pit=no
+  fi
+fi
+
+##########################################
+# test for KVM_CAP_DEVICE_ASSIGNMENT
+
+if test "$kvm_cap_device_assignment" != "no" ; then
+  if test "$kvm" = "no" -a "$kvm_cap_device_assignment" = "yes" ; then
+      feature_not_found "kvm_cap_device_assignment (kvm is not enabled)"
+  fi
+  cat > $TMPC <<EOF
+#include <linux/kvm.h>
+#ifndef KVM_CAP_DEVICE_ASSIGNMENT
+#error "kvm no device assignment capability"
+#endif
+int main(void) { return 0; }
+EOF
+  if compile_prog "$kvm_cflags" "" ; then
+    kvm_cap_device_assignment=yes
+  else
+    if test "$kvm_cap_device_assignment" = "yes" ; then
+      feature_not_found "kvm_cap_device_assigment"
+    fi
+    kvm_cap_device_assignment=no
+  fi
+fi
+
+##########################################
+# libpci probe for kvm_cap_device_assignment
+if test $kvm_cap_device_assignment = "yes" ; then
+  cat > $TMPC << EOF
+#include <pci/pci.h>
+#ifndef PCI_VENDOR_ID
+#error NO LIBPCI
+#endif
+int main(void) { struct pci_access a; pci_init(&a); return 0; }
+EOF
+  if compile_prog "" "-lpci -lz" ; then
+    libs_softmmu="-lpci -lz $libs_softmmu"
+  else
+    echo
+    echo "Error: libpci check failed"
+    echo "Disable KVM Device Assignment capability."
+    echo
+    kvm_cap_device_assignment=no
+  fi
+fi
+
 ##########################################
 # pthread probe
 PTHREADLIBS_LIST="-lpthread -lpthreadGC2"
@@ -1612,6 +1736,21 @@ if compile_prog "" "" ; then
   splice=yes
 fi
 
+##########################################
+# signalfd probe
+signalfd="no"
+cat > $TMPC << EOF
+#define _GNU_SOURCE
+#include <unistd.h>
+#include <sys/syscall.h>
+#include <signal.h>
+int main(void) { return syscall(SYS_signalfd, -1, NULL, _NSIG / 8); }
+EOF
+
+if $cc $ARCH_CFLAGS -o $TMPE $TMPC 2> /dev/null ; then
+  signalfd=yes
+fi
+
 # check if eventfd is supported
 eventfd=no
 cat > $TMPC << EOF
@@ -1792,8 +1931,9 @@ fi
 
 # Consult white-list to determine whether to enable werror
 # by default.  Only enable by default for git builds
+z_version=`cut -f3 -d. $source_path/VERSION`
+
 if test -z "$werror" ; then
-    z_version=`cut -f3 -d. $source_path/VERSION`
     if test "$z_version" = "50" -a \
         "$linux" = "yes" ; then
         werror="yes"
@@ -1802,6 +1942,16 @@ if test -z "$werror" ; then
     fi
 fi
 
+# Disable zero malloc errors for official releases unless explicitly told to
+# enable/disable
+if test -z "$zero_malloc" ; then
+    if test "$z_version" = "50" ; then
+	zero_malloc="no"
+    else
+	zero_malloc="yes"
+    fi
+fi
+
 if test "$werror" = "yes" ; then
     QEMU_CFLAGS="-Werror $QEMU_CFLAGS"
 fi
@@ -1830,6 +1980,20 @@ else
   binsuffix="/bin"
 fi
 
+if test -f kvm/kernel/configure; then
+    kvm_kmod="yes"
+    kmod_args=""
+    if test -n "$kerneldir"; then
+        kmod_args="--kerneldir=$kerneldir"
+    fi
+    if test "$kvm_trace" = "yes"; then
+        kmod_args="$kmod_args --with-kvm-trace"
+    fi
+    # hope there are no spaces in kmod_args; can't use arrays because of
+    # dash.
+    (cd kvm/kernel; ./configure $kmod_args)
+fi
+
 echo "Install prefix    $prefix"
 echo "BIOS directory    $prefix$datasuffix"
 echo "binary directory  $prefix$binsuffix"
@@ -1873,6 +2037,7 @@ if test -n "$sparc_cpu"; then
     echo "Target Sparc Arch $sparc_cpu"
 fi
 echo "xen support       $xen"
+echo "CPU emulation     $cpu_emulation"
 echo "brlapi support    $brlapi"
 echo "bluez  support    $bluez"
 echo "Documentation     $docs"
@@ -1886,6 +2051,9 @@ echo "IO thread         $io_thread"
 echo "Linux AIO support $linux_aio"
 echo "Install blobs     $blobs"
 echo "KVM support       $kvm"
+echo "KVM PIT support   $kvm_cap_pit"
+echo "KVM device assig. $kvm_cap_device_assignment"
+echo "KVM trace support $kvm_trace"
 echo "fdt support       $fdt"
 echo "preadv support    $preadv"
 echo "fdatasync         $fdatasync"
@@ -2092,6 +2260,9 @@ fi
 if test "$fdt" = "yes" ; then
   echo "CONFIG_FDT=y" >> $config_host_mak
 fi
+if test "$signalfd" = "yes" ; then
+  echo "CONFIG_SIGNALFD=y" >> $config_host_mak
+fi
 if test "$need_offsetof" = "yes" ; then
   echo "CONFIG_NEED_OFFSETOF=y" >> $config_host_mak
 fi
@@ -2101,6 +2272,11 @@ fi
 if test "$fdatasync" = "yes" ; then
   echo "CONFIG_FDATASYNC=y" >> $config_host_mak
 fi
+if test $cpu_emulation = "yes"; then
+  echo "CONFIG_CPU_EMULATION=y" >> $config_host_mak
+else
+  echo "CONFIG_NO_CPU_EMULATION=y" >> $config_host_mak
+fi
 
 # XXX: suppress that
 if [ "$bsd" = "yes" ] ; then
@@ -2109,6 +2285,10 @@ fi
 
 echo "CONFIG_UNAME_RELEASE=\"$uname_release\"" >> $config_host_mak
 
+if test "$zero_malloc" = "yes" ; then
+  echo "CONFIG_ZERO_MALLOC=y" >> $config_host_mak
+fi
+
 # USB host support
 case "$usb" in
 linux)
@@ -2122,6 +2302,8 @@ bsd)
 ;;
 esac
 
+echo "KVM_KMOD=$kvm_kmod" >> $config_host_mak
+
 tools=
 if test `expr "$target_list" : ".*softmmu.*"` != 0 ; then
   tools="qemu-img\$(EXESUF) qemu-io\$(EXESUF) $tools"
@@ -2272,6 +2454,9 @@ case "$target_arch2" in
     TARGET_BASE_ARCH=i386
     target_phys_bits=64
   ;;
+  ia64)
+    target_phys_bits=64
+  ;;
   alpha)
     target_phys_bits=64
   ;;
@@ -2402,6 +2587,12 @@ case "$target_arch2" in
       \( "$target_arch2" = "i386"   -a "$cpu" = "x86_64" \) \) ; then
       echo "CONFIG_KVM=y" >> $config_target_mak
       echo "KVM_CFLAGS=$kvm_cflags" >> $config_target_mak
+      if test $kvm_cap_pit = "yes" ; then
+        echo "CONFIG_KVM_PIT=y" >> $config_target_mak
+      fi
+      if test $kvm_cap_device_assignment = "yes" ; then
+        echo "CONFIG_KVM_DEVICE_ASSIGNMENT=y" >> $config_target_mak
+      fi
     fi
 esac
 echo "TARGET_PHYS_ADDR_BITS=$target_phys_bits" >> $config_target_mak
@@ -2610,7 +2801,7 @@ if test "$source_path_used" = "yes" ; then
 fi
 
 # temporary config to build submodules
-for rom in seabios vgabios ; do
+for rom in seabios vgabios; do
     config_mak=roms/$rom/config.mak
     echo "# Automatically generated by configure - do not modify" >> $config_mak
     echo "SRC_PATH=$source_path/roms/$rom" >> $config_mak
@@ -2636,3 +2827,6 @@ d=libuser
 mkdir -p $d
 rm -f $d/Makefile
 ln -s $source_path/Makefile.user $d/Makefile
+if test "$static" = "no" -a "$user_pie" = "yes" ; then
+  echo "QEMU_CFLAGS+=-fpie" > $d/config.mak
+fi

console.c

@@ -1384,6 +1384,16 @@ static void text_console_do_init(CharDriverState *chr, DisplayState *ds, QemuOpt
     s->t_attrib = s->t_attrib_default;
     text_console_resize(s);
 
+    if (chr->label) {
+        char msg[128];
+        int len;
+
+        s->t_attrib.bgcol = COLOR_BLUE;
+        len = snprintf(msg, sizeof(msg), "%s console\r\n", chr->label);
+        console_puts(chr, (uint8_t*)msg, len);
+        s->t_attrib = s->t_attrib_default;
+    }
+
     qemu_chr_generic_open(chr);
     if (chr->init)
         chr->init(chr);

console.h

@@ -44,7 +44,8 @@ struct MouseTransformInfo {
     int a[7];
 };
 
-void do_info_mice(Monitor *mon);
+void do_info_mice_print(Monitor *mon, const QObject *data);
+void do_info_mice(Monitor *mon, QObject **ret_data);
 void do_mouse_set(Monitor *mon, const QDict *qdict);
 
 /* keysym is a unicode code except for special keys (see QEMU_KEY_xxx
@@ -322,7 +323,8 @@ void vnc_display_init(DisplayState *ds);
 void vnc_display_close(DisplayState *ds);
 int vnc_display_open(DisplayState *ds, const char *display);
 int vnc_display_password(DisplayState *ds, const char *password);
-void do_info_vnc(Monitor *mon);
+void do_info_vnc_print(Monitor *mon, const QObject *data);
+void do_info_vnc(Monitor *mon, QObject **ret_data);
 char *vnc_display_local_addr(DisplayState *ds);
 
 /* curses.c */

cpu-all.h

@@ -849,6 +849,7 @@ extern int phys_ram_fd;
 extern uint8_t *phys_ram_dirty;
 extern ram_addr_t ram_size;
 extern ram_addr_t last_ram_offset;
+extern uint8_t *bios_mem;
 
 /* physical memory access */
 
@@ -1017,7 +1018,8 @@ static inline int64_t cpu_get_real_ticks (void)
 #endif
 }
 
-#elif (defined(__mips_isa_rev) && __mips_isa_rev >= 2) || defined(__linux__)
+#elif defined(__mips__) && \
+      ((defined(__mips_isa_rev) && __mips_isa_rev >= 2) || defined(__linux__))
 /*
  * binutils wants to use rdhwr only on mips32r2
  * but as linux kernel emulate it, it's fine

cpu-common.h

@@ -34,6 +34,7 @@ void qemu_ram_free(ram_addr_t addr);
 /* This should only be used for ram local to a device.  */
 void *qemu_get_ram_ptr(ram_addr_t addr);
 /* This should not be used by devices.  */
+int do_qemu_ram_addr_from_host(void *ptr, ram_addr_t *ram_addr);
 ram_addr_t qemu_ram_addr_from_host(void *ptr);
 
 int cpu_register_io_memory(CPUReadMemoryFunc * const *mem_read,

cpu-defs.h

@@ -27,6 +27,7 @@
 #include <setjmp.h>
 #include <inttypes.h>
 #include <signal.h>
+#include <pthread.h>
 #include "osdep.h"
 #include "qemu-queue.h"
 #include "targphys.h"
@@ -134,6 +135,16 @@ typedef struct CPUWatchpoint {
     QTAILQ_ENTRY(CPUWatchpoint) entry;
 } CPUWatchpoint;
 
+/* forward decleration */
+struct qemu_work_item;
+
+struct KVMCPUState {
+    pthread_t thread;
+    int signalled;
+    struct qemu_work_item *queued_work_first, *queued_work_last;
+    int regs_modified;
+};
+
 #define CPU_TEMP_BUF_NLONGS 128
 #define CPU_COMMON                                                      \
     struct TranslationBlock *current_tb; /* currently executing TB  */  \
@@ -146,8 +157,6 @@ typedef struct CPUWatchpoint {
     target_ulong mem_io_vaddr; /* target virtual addr at which the      \
                                      memory was accessed */             \
     uint32_t halted; /* Nonzero if the CPU is in suspend state */       \
-    uint32_t stop;   /* Stop request */                                 \
-    uint32_t stopped; /* Artificially stopped */                        \
     uint32_t interrupt_request;                                         \
     volatile sig_atomic_t exit_request;                                 \
     /* The meaning of the MMU modes is defined in the target code. */   \
@@ -188,6 +197,7 @@ typedef struct CPUWatchpoint {
     int nr_cores;  /* number of cores within this CPU package */        \
     int nr_threads;/* number of threads within this CPU */              \
     int running; /* Nonzero if cpu is currently running(usermode).  */  \
+    int thread_id;							\
     /* user data */                                                     \
     void *opaque;                                                       \
                                                                         \
@@ -197,6 +207,9 @@ typedef struct CPUWatchpoint {
     const char *cpu_model_str;                                          \
     struct KVMState *kvm_state;                                         \
     struct kvm_run *kvm_run;                                            \
-    int kvm_fd;
+    int kvm_fd;                                                         \
+    uint32_t stop;   /* Stop request */                                 \
+    uint32_t stopped; /* Artificially stopped */                        \
+    struct KVMCPUState kvm_cpu_state;
 
 #endif

cpu-exec.c

@@ -19,7 +19,9 @@
 #include "config.h"
 #include "exec.h"
 #include "disas.h"
+#if !defined(TARGET_IA64)
 #include "tcg.h"
+#endif
 #include "kvm.h"
 
 #if !defined(CONFIG_SOFTMMU)
@@ -38,6 +40,8 @@
 #endif
 #endif
 
+#include "qemu-kvm.h"
+
 #if defined(__sparc__) && !defined(CONFIG_SOLARIS)
 // Work around ugly bugs in glibc that mangle global register contents
 #undef env
@@ -232,11 +236,13 @@ int cpu_exec(CPUState *env1)
 
     env_to_regs();
 #if defined(TARGET_I386)
-    /* put eflags in CPU temporary format */
+    if (!kvm_enabled()) {
-    CC_SRC = env->eflags & (CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
+        /* put eflags in CPU temporary format */
-    DF = 1 - (2 * ((env->eflags >> 10) & 1));
+        CC_SRC = env->eflags & (CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
-    CC_OP = CC_OP_EFLAGS;
+        DF = 1 - (2 * ((env->eflags >> 10) & 1));
-    env->eflags &= ~(DF_MASK | CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
+        CC_OP = CC_OP_EFLAGS;
+        env->eflags &= ~(DF_MASK | CC_O | CC_S | CC_Z | CC_A | CC_P | CC_C);
+    }
 #elif defined(TARGET_SPARC)
 #elif defined(TARGET_M68K)
     env->cc_op = CC_OP_FLAGS;
@@ -250,6 +256,7 @@ int cpu_exec(CPUState *env1)
 #elif defined(TARGET_SH4)
 #elif defined(TARGET_CRIS)
 #elif defined(TARGET_S390X)
+#elif defined(TARGET_IA64)
     /* XXXXX */
 #else
 #error unsupported target CPU
@@ -317,6 +324,8 @@ int cpu_exec(CPUState *env1)
                     do_interrupt(env);
 #elif defined(TARGET_M68K)
                     do_interrupt(0);
+#elif defined(TARGET_IA64)
+		    do_interrupt(env);
 #endif
 #endif
                 }
@@ -672,6 +681,7 @@ int cpu_exec(CPUState *env1)
 #elif defined(TARGET_MICROBLAZE)
 #elif defined(TARGET_MIPS)
 #elif defined(TARGET_SH4)
+#elif defined(TARGET_IA64)
 #elif defined(TARGET_ALPHA)
 #elif defined(TARGET_CRIS)
 #elif defined(TARGET_S390X)

cutils.c

@@ -218,6 +218,11 @@ void qemu_iovec_to_buffer(QEMUIOVector *qiov, void *buf)
     }
 }
 
+/*
+ * No dma flushing needed here, as the aio code will call dma_bdrv_cb()
+ * on completion as well, which will result in a call to
+ * dma_bdrv_unmap() which will do the flushing ....
+ */
 void qemu_iovec_from_buffer(QEMUIOVector *qiov, const void *buf, size_t count)
 {
     const uint8_t *p = (const uint8_t *)buf;

default-configs/sh4-softmmu.mak

@@ -2,3 +2,4 @@
 
 CONFIG_USB_OHCI=y
 CONFIG_PTIMER=y
+CONFIG_ISA_MMIO=y

default-configs/sh4eb-softmmu.mak

@@ -2,3 +2,4 @@
 
 CONFIG_USB_OHCI=y
 CONFIG_PTIMER=y
+CONFIG_ISA_MMIO=y

dma-helpers.c

@@ -160,6 +160,10 @@ static BlockDriverAIOCB *dma_bdrv_io(
     dbs->is_write = is_write;
     dbs->bh = NULL;
     qemu_iovec_init(&dbs->iov, sg->nsg);
+    /*
+     * DMA flushing is handled in dma_bdrv_cb() calling dma_bdrv_unmap()
+     * so we don't need to do that here.
+     */
     dma_bdrv_cb(dbs, 0);
     if (!dbs->acb) {
         qemu_aio_release(dbs);

dyngen-exec.h

@@ -69,9 +69,9 @@ extern int printf(const char *, ...);
 #define AREG1 "r14"
 #define AREG2 "r15"
 #elif defined(__mips__)
-#define AREG0 "fp"
+#define AREG0 "s0"
-#define AREG1 "s0"
+#define AREG1 "s1"
-#define AREG2 "s1"
+#define AREG2 "fp"
 #elif defined(__sparc__)
 #ifdef CONFIG_SOLARIS
 #define AREG0 "g2"

exec.c

@@ -34,7 +34,13 @@
 #include "cpu.h"
 #include "exec-all.h"
 #include "qemu-common.h"
+#include "cache-utils.h"
+
+#if !defined(TARGET_IA64)
 #include "tcg.h"
+#endif
+#include "qemu-kvm.h"
+
 #include "hw/hw.h"
 #include "osdep.h"
 #include "kvm.h"
@@ -74,6 +80,8 @@
 #define TARGET_PHYS_ADDR_SPACE_BITS 42
 #elif defined(TARGET_I386)
 #define TARGET_PHYS_ADDR_SPACE_BITS 36
+#elif defined(TARGET_IA64)
+#define TARGET_PHYS_ADDR_SPACE_BITS 36
 #else
 #define TARGET_PHYS_ADDR_SPACE_BITS 32
 #endif
@@ -111,6 +119,7 @@ uint8_t *code_gen_ptr;
 #if !defined(CONFIG_USER_ONLY)
 int phys_ram_fd;
 uint8_t *phys_ram_dirty;
+uint8_t *bios_mem;
 static int in_migration;
 
 typedef struct RAMBlock {
@@ -412,6 +421,9 @@ static uint8_t static_code_gen_buffer[DEFAULT_CODE_GEN_BUFFER_SIZE];
 
 static void code_gen_alloc(unsigned long tb_size)
 {
+    if (kvm_enabled())
+        return;
+
 #ifdef USE_STATIC_CODE_GEN_BUFFER
     code_gen_buffer = static_code_gen_buffer;
     code_gen_buffer_size = DEFAULT_CODE_GEN_BUFFER_SIZE;
@@ -588,6 +600,11 @@ void cpu_exec_init(CPUState *env)
     env->numa_node = 0;
     QTAILQ_INIT(&env->breakpoints);
     QTAILQ_INIT(&env->watchpoints);
+#ifdef __WIN32
+    env->thread_id = GetCurrentProcessId();
+#else
+    env->thread_id = getpid();
+#endif
     *penv = env;
 #if defined(CONFIG_USER_ONLY)
     cpu_list_unlock();
@@ -1553,6 +1570,8 @@ void cpu_interrupt(CPUState *env, int mask)
 
     old_mask = env->interrupt_request;
     env->interrupt_request |= mask;
+    if (kvm_enabled() && !kvm_irqchip_in_kernel())
+	kvm_update_interrupt_request(env);
 
 #ifndef CONFIG_USER_ONLY
     /*
@@ -1880,7 +1899,6 @@ void cpu_physical_memory_reset_dirty(ram_addr_t start, ram_addr_t end,
 
 int cpu_physical_memory_set_dirty_tracking(int enable)
 {
-    in_migration = enable;
     if (kvm_enabled()) {
         return kvm_set_migration_log(enable);
     }
@@ -2404,6 +2422,113 @@ void qemu_unregister_coalesced_mmio(target_phys_addr_t addr, ram_addr_t size)
         kvm_uncoalesce_mmio_region(addr, size);
 }
 
+#ifdef __linux__
+
+#include <sys/vfs.h>
+
+#define HUGETLBFS_MAGIC       0x958458f6
+
+static long gethugepagesize(const char *path)
+{
+    struct statfs fs;
+    int ret;
+
+    do {
+	    ret = statfs(path, &fs);
+    } while (ret != 0 && errno == EINTR);
+
+    if (ret != 0) {
+	    perror("statfs");
+	    return 0;
+    }
+
+    if (fs.f_type != HUGETLBFS_MAGIC)
+	    fprintf(stderr, "Warning: path not on HugeTLBFS: %s\n", path);
+
+    return fs.f_bsize;
+}
+
+static void *file_ram_alloc(ram_addr_t memory, const char *path)
+{
+    char *filename;
+    void *area;
+    int fd;
+#ifdef MAP_POPULATE
+    int flags;
+#endif
+    unsigned long hpagesize;
+    extern int mem_prealloc;
+
+    if (!path) {
+        return NULL;
+    }
+
+    hpagesize = gethugepagesize(path);
+    if (!hpagesize) {
+	return NULL;
+    }
+
+    if (memory < hpagesize) {
+        return NULL;
+    }
+
+    if (kvm_enabled() && !kvm_has_sync_mmu()) {
+        fprintf(stderr, "host lacks mmu notifiers, disabling --mem-path\n");
+        return NULL;
+    }
+
+    if (asprintf(&filename, "%s/kvm.XXXXXX", path) == -1) {
+	return NULL;
+    }
+
+    fd = mkstemp(filename);
+    if (fd < 0) {
+	perror("mkstemp");
+	free(filename);
+	return NULL;
+    }
+    unlink(filename);
+    free(filename);
+
+    memory = (memory+hpagesize-1) & ~(hpagesize-1);
+
+    /*
+     * ftruncate is not supported by hugetlbfs in older
+     * hosts, so don't bother checking for errors.
+     * If anything goes wrong with it under other filesystems,
+     * mmap will fail.
+     */
+    ftruncate(fd, memory);
+
+#ifdef MAP_POPULATE
+    /* NB: MAP_POPULATE won't exhaustively alloc all phys pages in the case
+     * MAP_PRIVATE is requested.  For mem_prealloc we mmap as MAP_SHARED
+     * to sidestep this quirk.
+     */
+    flags = mem_prealloc ? MAP_POPULATE|MAP_SHARED : MAP_PRIVATE;
+    area = mmap(0, memory, PROT_READ|PROT_WRITE, flags, fd, 0);
+#else
+    area = mmap(0, memory, PROT_READ|PROT_WRITE, MAP_PRIVATE, fd, 0);
+#endif
+    if (area == MAP_FAILED) {
+	perror("alloc_mem_area: can't mmap hugetlbfs pages");
+	close(fd);
+	return (NULL);
+    }
+    return area;
+}
+
+#else
+
+static void *file_ram_alloc(ram_addr_t memory, const char *path)
+{
+    return NULL;
+}
+
+#endif
+
+extern const char *mem_path;
+
 ram_addr_t qemu_ram_alloc(ram_addr_t size)
 {
     RAMBlock *new_block;
@@ -2411,16 +2536,20 @@ ram_addr_t qemu_ram_alloc(ram_addr_t size)
     size = TARGET_PAGE_ALIGN(size);
     new_block = qemu_malloc(sizeof(*new_block));
 
+    new_block->host = file_ram_alloc(size, mem_path);
+    if (!new_block->host) {
 #if defined(TARGET_S390X) && defined(CONFIG_KVM)
     /* XXX S390 KVM requires the topmost vma of the RAM to be < 256GB */
-    new_block->host = mmap((void*)0x1000000, size, PROT_EXEC|PROT_READ|PROT_WRITE,
+        new_block->host = mmap((void*)0x1000000, size,
-                           MAP_SHARED | MAP_ANONYMOUS, -1, 0);
+                               PROT_EXEC|PROT_READ|PROT_WRITE,
+                               MAP_SHARED | MAP_ANONYMOUS, -1, 0);
 #else
-    new_block->host = qemu_vmalloc(size);
+        new_block->host = qemu_vmalloc(size);
 #endif
 #ifdef MADV_MERGEABLE
-    madvise(new_block->host, size, MADV_MERGEABLE);
+        madvise(new_block->host, size, MADV_MERGEABLE);
 #endif
+    }
     new_block->offset = last_ram_offset;
     new_block->length = size;
 
@@ -2482,9 +2611,7 @@ void *qemu_get_ram_ptr(ram_addr_t addr)
     return block->host + (addr - block->offset);
 }
 
-/* Some of the softmmu routines need to translate from a host pointer
+int do_qemu_ram_addr_from_host(void *ptr, ram_addr_t *ram_addr)
-   (typically a TLB entry) back to a ram offset.  */
-ram_addr_t qemu_ram_addr_from_host(void *ptr)
 {
     RAMBlock *prev;
     RAMBlock **prevp;
@@ -2501,11 +2628,23 @@ ram_addr_t qemu_ram_addr_from_host(void *ptr)
         prev = block;
         block = block->next;
     }
-    if (!block) {
+    if (!block)
+        return -1;
+    *ram_addr = block->offset + (host - block->host);
+    return 0;
+}
+
+/* Some of the softmmu routines need to translate from a host pointer
+   (typically a TLB entry) back to a ram offset.  */
+ram_addr_t qemu_ram_addr_from_host(void *ptr)
+{
+    ram_addr_t ram_addr;
+
+    if (do_qemu_ram_addr_from_host(ptr, &ram_addr)) {
         fprintf(stderr, "Bad ram pointer %p\n", ptr);
         abort();
     }
-    return block->offset + (host - block->host);
+    return ram_addr;
 }
 
 static uint32_t unassigned_mem_readb(void *opaque, target_phys_addr_t addr)
@@ -3091,6 +3230,11 @@ void cpu_physical_memory_rw(target_phys_addr_t addr, uint8_t *buf,
                     phys_ram_dirty[addr1 >> TARGET_PAGE_BITS] |=
                         (0xff & ~CODE_DIRTY_FLAG);
                 }
+		/* qemu doesn't execute guest code directly, but kvm does
+		   therefore flush instruction caches */
+		if (kvm_enabled())
+		    flush_icache_range((unsigned long)ptr,
+				       ((unsigned long)ptr)+l);
             }
         } else {
             if ((pd & ~TARGET_PAGE_MASK) > IO_MEM_ROM &&
@@ -3283,6 +3427,8 @@ void *cpu_physical_memory_map(target_phys_addr_t addr,
 void cpu_physical_memory_unmap(void *buffer, target_phys_addr_t len,
                                int is_write, target_phys_addr_t access_len)
 {
+    unsigned long flush_len = (unsigned long)access_len;
+
     if (buffer != bounce.buffer) {
         if (is_write) {
             ram_addr_t addr1 = qemu_ram_addr_from_host(buffer);
@@ -3300,14 +3446,16 @@ void cpu_physical_memory_unmap(void *buffer, target_phys_addr_t len,
                 }
                 addr1 += l;
                 access_len -= l;
-            }
+	    }
+	    dma_flush_range((unsigned long)buffer,
+			    (unsigned long)buffer + flush_len);
         }
         return;
     }
     if (is_write) {
         cpu_physical_memory_write(bounce.addr, bounce.buffer, access_len);
     }
-    qemu_free(bounce.buffer);
+    qemu_vfree(bounce.buffer);
     bounce.buffer = NULL;
     cpu_notify_map_clients();
 }
@@ -3668,7 +3816,9 @@ void dump_exec_info(FILE *f,
     cpu_fprintf(f, "TB flush count      %d\n", tb_flush_count);
     cpu_fprintf(f, "TB invalidate count %d\n", tb_phys_invalidate_count);
     cpu_fprintf(f, "TLB flush count     %d\n", tlb_flush_count);
+#ifdef CONFIG_PROFILER
     tcg_dump_info(f, cpu_fprintf);
+#endif
 }
 
 #if !defined(CONFIG_USER_ONLY)

fpu/softfloat-native.c

@@ -5,6 +5,7 @@
 #if defined(CONFIG_SOLARIS)
 #include <fenv.h>
 #endif
+#include "config-host.h"
 
 void set_float_rounding_mode(int val STATUS_PARAM)
 {

gdbstub.c

@@ -34,6 +34,7 @@
 #include "sysemu.h"
 #include "gdbstub.h"
 #endif
+#include "qemu-kvm.h"
 
 #define MAX_PACKET_LENGTH 4096
 

host-utils.h

@@ -164,7 +164,7 @@ static inline int ctz64(uint64_t val)
 {
 #if QEMU_GNUC_PREREQ(3, 4)
     if (val)
-        return __builtin_ctz(val);
+        return __builtin_ctzll(val);
     else
         return 64;
 #else

hw/acpi.c

@@ -23,6 +23,8 @@
 #include "i2c.h"
 #include "smbus.h"
 #include "kvm.h"
+#include "qemu-kvm.h"
+#include "string.h"
 
 //#define DEBUG
 
@@ -521,6 +523,13 @@ i2c_bus *piix4_pm_init(PCIBus *bus, int devfn, uint32_t smb_io_base,
 
     pci_conf[0x40] = 0x01; /* PM io base read only bit */
 
+#if defined(TARGET_IA64)
+    pci_conf[0x40] = 0x41; /* PM io base read only bit */
+    pci_conf[0x41] = 0x1f;
+    pm_write_config(s, 0x80, 0x01, 1); /*Set default pm_io_base 0x1f40*/
+    s->pmcntrl = SCI_EN;
+#endif
+
     register_ioport_write(0xb2, 2, 1, pm_smi_writeb, s);
     register_ioport_read(0xb2, 2, 1, pm_smi_readb, s);
 
@@ -559,12 +568,14 @@ i2c_bus *piix4_pm_init(PCIBus *bus, int devfn, uint32_t smb_io_base,
 }
 
 #define GPE_BASE 0xafe0
+#define PROC_BASE 0xaf00
 #define PCI_BASE 0xae00
 #define PCI_EJ_BASE 0xae08
 
 struct gpe_regs {
     uint16_t sts; /* status */
     uint16_t en;  /* enabled */
+    uint8_t cpus_sts[32];
 };
 
 struct pci_status {
@@ -587,6 +598,10 @@ static uint32_t gpe_readb(void *opaque, uint32_t addr)
     uint32_t val = 0;
     struct gpe_regs *g = opaque;
     switch (addr) {
+        case PROC_BASE ... PROC_BASE+31:
+            val = g->cpus_sts[addr - PROC_BASE];
+            break;
+
         case GPE_BASE:
         case GPE_BASE + 1:
             val = gpe_read_val(g->sts, addr);
@@ -629,6 +644,10 @@ static void gpe_writeb(void *opaque, uint32_t addr, uint32_t val)
 {
     struct gpe_regs *g = opaque;
     switch (addr) {
+        case PROC_BASE ... PROC_BASE + 31:
+            /* don't allow to change cpus_sts from inside a guest */
+            break;
+
         case GPE_BASE:
         case GPE_BASE + 1:
             gpe_reset_val(&g->sts, addr, val);
@@ -712,22 +731,72 @@ static void pciej_write(void *opaque, uint32_t addr, uint32_t val)
 #endif
 }
 
+static const char *model;
+
 static int piix4_device_hotplug(PCIDevice *dev, int state);
 
-void piix4_acpi_system_hot_add_init(PCIBus *bus)
+void piix4_acpi_system_hot_add_init(PCIBus *bus, const char *cpu_model)
 {
+    int i = 0, cpus = smp_cpus;
+
+    while (cpus > 0) {
+        gpe.cpus_sts[i++] = (cpus < 8) ? (1 << cpus) - 1 : 0xff;
+        cpus -= 8;
+    }
     register_ioport_write(GPE_BASE, 4, 1, gpe_writeb, &gpe);
     register_ioport_read(GPE_BASE, 4, 1,  gpe_readb, &gpe);
 
+    register_ioport_write(PROC_BASE, 32, 1, gpe_writeb, &gpe);
+    register_ioport_read(PROC_BASE, 32, 1,  gpe_readb, &gpe);
+
     register_ioport_write(PCI_BASE, 8, 4, pcihotplug_write, &pci0_status);
     register_ioport_read(PCI_BASE, 8, 4,  pcihotplug_read, &pci0_status);
 
     register_ioport_write(PCI_EJ_BASE, 4, 4, pciej_write, bus);
     register_ioport_read(PCI_EJ_BASE, 4, 4,  pciej_read, bus);
 
+    model = cpu_model;
+
     pci_bus_hotplug(bus, piix4_device_hotplug);
 }
 
+#if defined(TARGET_I386)
+static void enable_processor(struct gpe_regs *g, int cpu)
+{
+    g->sts |= 4;
+    g->cpus_sts[cpu/8] |= (1 << (cpu%8));
+}
+
+static void disable_processor(struct gpe_regs *g, int cpu)
+{
+    g->sts |= 4;
+    g->cpus_sts[cpu/8] &= ~(1 << (cpu%8));
+}
+
+void qemu_system_cpu_hot_add(int cpu, int state)
+{
+    CPUState *env;
+
+    if (state && !qemu_get_cpu(cpu)) {
+        env = pc_new_cpu(model);
+        if (!env) {
+            fprintf(stderr, "cpu %d creation failed\n", cpu);
+            return;
+        }
+        env->cpuid_apic_id = cpu;
+    }
+
+    if (state)
+        enable_processor(&gpe, cpu);
+    else
+        disable_processor(&gpe, cpu);
+    if (gpe.en & 4) {
+        qemu_set_irq(pm_state->irq, 1);
+        qemu_set_irq(pm_state->irq, 0);
+    }
+}
+#endif
+
 static void enable_device(struct pci_status *p, struct gpe_regs *g, int slot)
 {
     g->sts |= 2;

hw/apic.c

@@ -24,6 +24,8 @@
 #include "host-utils.h"
 #include "kvm.h"
 
+#include "qemu-kvm.h"
+
 //#define DEBUG_APIC
 
 /* APIC Local Vector Table */
@@ -299,8 +301,11 @@ void cpu_set_apic_base(CPUState *env, uint64_t val)
 #endif
     if (!s)
         return;
-    s->apicbase = (val & 0xfffff000) |
+    if (kvm_enabled() && kvm_irqchip_in_kernel())
-        (s->apicbase & (MSR_IA32_APICBASE_BSP | MSR_IA32_APICBASE_ENABLE));
+        s->apicbase = val;
+    else
+        s->apicbase = (val & 0xfffff000) |
+            (s->apicbase & (MSR_IA32_APICBASE_BSP | MSR_IA32_APICBASE_ENABLE));
     /* if disabled, cannot be enabled again */
     if (!(val & MSR_IA32_APICBASE_ENABLE)) {
         s->apicbase &= ~MSR_IA32_APICBASE_ENABLE;
@@ -393,6 +398,11 @@ int apic_get_irq_delivered(void)
     return apic_irq_delivered;
 }
 
+void apic_set_irq_delivered(void)
+{
+    apic_irq_delivered = 1;
+}
+
 static void apic_set_irq(APICState *s, int vector_num, int trigger_mode)
 {
     apic_irq_delivered += !get_bit(s->irr, vector_num);
@@ -478,6 +488,7 @@ void apic_init_reset(CPUState *env)
     if (!s)
         return;
 
+    cpu_synchronize_state(env);
     s->tpr = 0;
     s->spurious_vec = 0xff;
     s->log_dest = 0;
@@ -497,6 +508,13 @@ void apic_init_reset(CPUState *env)
     s->wait_for_sipi = 1;
 
     env->halted = !(s->apicbase & MSR_IA32_APICBASE_BSP);
+#ifdef KVM_CAP_MP_STATE
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        env->mp_state
+            = env->halted ? KVM_MP_STATE_UNINITIALIZED : KVM_MP_STATE_RUNNABLE;
+        kvm_load_mpstate(env);
+    }
+#endif
 }
 
 static void apic_startup(APICState *s, int vector_num)
@@ -864,6 +882,115 @@ static void apic_mem_writel(void *opaque, target_phys_addr_t addr, uint32_t val)
     }
 }
 
+#ifdef KVM_CAP_IRQCHIP
+
+static inline uint32_t kapic_reg(struct kvm_lapic_state *kapic, int reg_id)
+{
+    return *((uint32_t *) (kapic->regs + (reg_id << 4)));
+}
+
+static inline void kapic_set_reg(struct kvm_lapic_state *kapic,
+                                 int reg_id, uint32_t val)
+{
+    *((uint32_t *) (kapic->regs + (reg_id << 4))) = val;
+}
+
+static void kvm_kernel_lapic_save_to_user(APICState *s)
+{
+    struct kvm_lapic_state apic;
+    struct kvm_lapic_state *kapic = &apic;
+    int i, v;
+
+    kvm_get_lapic(s->cpu_env, kapic);
+
+    s->id = kapic_reg(kapic, 0x2) >> 24;
+    s->tpr = kapic_reg(kapic, 0x8);
+    s->arb_id = kapic_reg(kapic, 0x9);
+    s->log_dest = kapic_reg(kapic, 0xd) >> 24;
+    s->dest_mode = kapic_reg(kapic, 0xe) >> 28;
+    s->spurious_vec = kapic_reg(kapic, 0xf);
+    for (i = 0; i < 8; i++) {
+        s->isr[i] = kapic_reg(kapic, 0x10 + i);
+        s->tmr[i] = kapic_reg(kapic, 0x18 + i);
+        s->irr[i] = kapic_reg(kapic, 0x20 + i);
+    }
+    s->esr = kapic_reg(kapic, 0x28);
+    s->icr[0] = kapic_reg(kapic, 0x30);
+    s->icr[1] = kapic_reg(kapic, 0x31);
+    for (i = 0; i < APIC_LVT_NB; i++)
+	s->lvt[i] = kapic_reg(kapic, 0x32 + i);
+    s->initial_count = kapic_reg(kapic, 0x38);
+    s->divide_conf = kapic_reg(kapic, 0x3e);
+
+    v = (s->divide_conf & 3) | ((s->divide_conf >> 1) & 4);
+    s->count_shift = (v + 1) & 7;
+
+    s->initial_count_load_time = qemu_get_clock(vm_clock);
+    apic_timer_update(s, s->initial_count_load_time);
+}
+
+static void kvm_kernel_lapic_load_from_user(APICState *s)
+{
+    struct kvm_lapic_state apic;
+    struct kvm_lapic_state *klapic = &apic;
+    int i;
+
+    memset(klapic, 0, sizeof apic);
+    kapic_set_reg(klapic, 0x2, s->id << 24);
+    kapic_set_reg(klapic, 0x8, s->tpr);
+    kapic_set_reg(klapic, 0xd, s->log_dest << 24);
+    kapic_set_reg(klapic, 0xe, s->dest_mode << 28 | 0x0fffffff);
+    kapic_set_reg(klapic, 0xf, s->spurious_vec);
+    for (i = 0; i < 8; i++) {
+        kapic_set_reg(klapic, 0x10 + i, s->isr[i]);
+        kapic_set_reg(klapic, 0x18 + i, s->tmr[i]);
+        kapic_set_reg(klapic, 0x20 + i, s->irr[i]);
+    }
+    kapic_set_reg(klapic, 0x28, s->esr);
+    kapic_set_reg(klapic, 0x30, s->icr[0]);
+    kapic_set_reg(klapic, 0x31, s->icr[1]);
+    for (i = 0; i < APIC_LVT_NB; i++)
+        kapic_set_reg(klapic, 0x32 + i, s->lvt[i]);
+    kapic_set_reg(klapic, 0x38, s->initial_count);
+    kapic_set_reg(klapic, 0x3e, s->divide_conf);
+
+    kvm_set_lapic(s->cpu_env, klapic);
+}
+
+#endif
+
+void qemu_kvm_load_lapic(CPUState *env)
+{
+#ifdef KVM_CAP_IRQCHIP
+    if (kvm_enabled() && kvm_vcpu_inited(env) && kvm_irqchip_in_kernel()) {
+        kvm_kernel_lapic_load_from_user(env->apic_state);
+    }
+#endif
+}
+
+static void apic_pre_save(void *opaque)
+{
+#ifdef KVM_CAP_IRQCHIP
+    APICState *s = (void *)opaque;
+
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_lapic_save_to_user(s);
+    }
+#endif
+}
+
+static int apic_post_load(void *opaque, int version_id)
+{
+#ifdef KVM_CAP_IRQCHIP
+    APICState *s = opaque;
+
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_lapic_load_from_user(s);
+    }
+#endif
+    return 0;
+}
+
 /* This function is only used for old state version 1 and 2 */
 static int apic_load_old(QEMUFile *f, void *opaque, int version_id)
 {
@@ -900,6 +1027,9 @@ static int apic_load_old(QEMUFile *f, void *opaque, int version_id)
 
     if (version_id >= 2)
         qemu_get_timer(f, s->timer);
+
+    qemu_kvm_load_lapic(s->cpu_env);
+
     return 0;
 }
 
@@ -930,7 +1060,9 @@ static const VMStateDescription vmstate_apic = {
         VMSTATE_INT64(next_time, APICState),
         VMSTATE_TIMER(timer, APICState),
         VMSTATE_END_OF_LIST()
-    }
+    },
+    .pre_save = apic_pre_save,
+    .post_load = apic_post_load,
 };
 
 static void apic_reset(void *opaque)
@@ -955,6 +1087,7 @@ static void apic_reset(void *opaque)
          */
         s->lvt[APIC_LVT_LINT0] = 0x700;
     }
+    qemu_kvm_load_lapic(s->cpu_env);
 }
 
 static CPUReadMemoryFunc * const apic_mem_read[3] = {
@@ -998,6 +1131,11 @@ int apic_init(CPUState *env)
     vmstate_register(s->idx, &vmstate_apic, s);
     qemu_register_reset(apic_reset, s);
 
+    /* apic_reset must be called before the vcpu threads are initialized and load
+     * registers, in qemu-kvm.
+     */
+    apic_reset(s);
+
     local_apics[s->idx] = s;
     return 0;
 }

hw/arm_timer.c

@@ -71,7 +71,7 @@ static void arm_timer_recalibrate(arm_timer_state *s, int reload)
 {
     uint32_t limit;
 
-    if ((s->control & TIMER_CTRL_PERIODIC) == 0) {
+    if ((s->control & (TIMER_CTRL_PERIODIC | TIMER_CTRL_ONESHOT)) == 0) {
         /* Free running.  */
         if (s->control & TIMER_CTRL_32BIT)
             limit = 0xffffffff;
@@ -113,7 +113,7 @@ static void arm_timer_write(void *opaque, target_phys_addr_t offset,
         case 1: freq >>= 4; break;
         case 2: freq >>= 8; break;
         }
-        arm_timer_recalibrate(s, 0);
+        arm_timer_recalibrate(s, s->control & TIMER_CTRL_ENABLE);
         ptimer_set_freq(s->timer, freq);
         if (s->control & TIMER_CTRL_ENABLE) {
             /* Restart the timer if still enabled.  */

hw/boards.h

@@ -19,8 +19,15 @@ typedef struct QEMUMachine {
     QEMUMachineInitFunc *init;
     int use_scsi;
     int max_cpus;
+    int no_serial:1,
+        no_parallel:1,
+        use_virtcon:1,
+        no_vga:1,
+        no_floppy:1,
+        no_cdrom:1,
+        no_sdcard:1;
     int is_default;
-    CompatProperty *compat_props;
+    GlobalProperty *compat_props;
     struct QEMUMachine *next;
 } QEMUMachine;
 

hw/bt-hid.c

@@ -552,7 +552,7 @@ static struct bt_device_s *bt_hid_init(struct bt_scatternet_s *net,
                     BT_HID_MTU, bt_hid_new_interrupt_ch);
 
     s->usbdev = dev;
-    s->btdev.device.lmp_name = s->usbdev->devname;
+    s->btdev.device.lmp_name = s->usbdev->product_desc;
     usb_hid_datain_cb(s->usbdev, s, bt_hid_datain);
 
     s->btdev.device.handle_destroy = bt_hid_destroy;
@@ -566,6 +566,6 @@ static struct bt_device_s *bt_hid_init(struct bt_scatternet_s *net,
 
 struct bt_device_s *bt_keyboard_init(struct bt_scatternet_s *net)
 {
-    USBDevice *dev = usb_create_simple(NULL /* FIXME */, "QEMU USB Keyboard");
+    USBDevice *dev = usb_create_simple(NULL /* FIXME */, "usb-kbd");
     return bt_hid_init(net, dev, class_keyboard);
 }

hw/cirrus_vga.c

@@ -32,6 +32,7 @@
 #include "console.h"
 #include "vga_int.h"
 #include "kvm.h"
+#include "qemu-kvm.h"
 #include "loader.h"
 
 /*
@@ -2552,6 +2553,7 @@ static CPUWriteMemoryFunc * const cirrus_linear_bitblt_write[3] = {
 
 static void map_linear_vram(CirrusVGAState *s)
 {
+    vga_dirty_log_stop(&s->vga);
     if (!s->vga.map_addr && s->vga.lfb_addr && s->vga.lfb_end) {
         s->vga.map_addr = s->vga.lfb_addr;
         s->vga.map_end = s->vga.lfb_end;
@@ -2561,13 +2563,19 @@ static void map_linear_vram(CirrusVGAState *s)
     if (!s->vga.map_addr)
         return;
 
+#ifndef TARGET_IA64
     s->vga.lfb_vram_mapped = 0;
 
+    cpu_register_physical_memory(isa_mem_base + 0xa0000, 0x8000,
+                                (s->vga.vram_offset + s->cirrus_bank_base[0]) | IO_MEM_UNASSIGNED);
+    cpu_register_physical_memory(isa_mem_base + 0xa8000, 0x8000,
+                                (s->vga.vram_offset + s->cirrus_bank_base[1]) | IO_MEM_UNASSIGNED);
     if (!(s->cirrus_srcptr != s->cirrus_srcptr_end)
         && !((s->vga.sr[0x07] & 0x01) == 0)
         && !((s->vga.gr[0x0B] & 0x14) == 0x14)
         && !(s->vga.gr[0x0B] & 0x02)) {
 
+        vga_dirty_log_stop(&s->vga);
         cpu_register_physical_memory(isa_mem_base + 0xa0000, 0x8000,
                                     (s->vga.vram_offset + s->cirrus_bank_base[0]) | IO_MEM_RAM);
         cpu_register_physical_memory(isa_mem_base + 0xa8000, 0x8000,
@@ -2579,17 +2587,23 @@ static void map_linear_vram(CirrusVGAState *s)
         cpu_register_physical_memory(isa_mem_base + 0xa0000, 0x20000,
                                      s->vga.vga_io_memory);
     }
+#endif
 
     vga_dirty_log_start(&s->vga);
 }
 
 static void unmap_linear_vram(CirrusVGAState *s)
 {
-    if (s->vga.map_addr && s->vga.lfb_addr && s->vga.lfb_end)
+    vga_dirty_log_stop(&s->vga);
+    if (s->vga.map_addr && s->vga.lfb_addr && s->vga.lfb_end) {
         s->vga.map_addr = s->vga.map_end = 0;
-
+         cpu_register_physical_memory(s->vga.lfb_addr, s->vga.vram_size,
+                                      s->cirrus_linear_io_addr);
+    }
     cpu_register_physical_memory(isa_mem_base + 0xa0000, 0x20000,
                                  s->vga.vga_io_memory);
+
+    vga_dirty_log_start(&s->vga);
 }
 
 /* Compute the memory access functions */
@@ -3143,6 +3157,8 @@ static void cirrus_pci_lfb_map(PCIDevice *d, int region_num,
 {
     CirrusVGAState *s = &DO_UPCAST(PCICirrusVGAState, dev, d)->cirrus_vga;
 
+    vga_dirty_log_stop(&s->vga);
+
     /* XXX: add byte swapping apertures */
     cpu_register_physical_memory(addr, s->vga.vram_size,
 				 s->cirrus_linear_io_addr);
@@ -3174,10 +3190,14 @@ static void pci_cirrus_write_config(PCIDevice *d,
     PCICirrusVGAState *pvs = DO_UPCAST(PCICirrusVGAState, dev, d);
     CirrusVGAState *s = &pvs->cirrus_vga;
 
+    vga_dirty_log_stop(&s->vga);
+
     pci_default_write_config(d, address, val, len);
     if (s->vga.map_addr && d->io_regions[0].addr == PCI_BAR_UNMAPPED)
         s->vga.map_addr = 0;
     cirrus_update_memory_access(s);
+
+    vga_dirty_log_start(&s->vga);
 }
 
 static int pci_cirrus_vga_initfn(PCIDevice *dev)
@@ -3209,22 +3229,21 @@ static int pci_cirrus_vga_initfn(PCIDevice *dev)
          pci_register_bar((PCIDevice *)d, 1, CIRRUS_PNPMMIO_SIZE,
                           PCI_BASE_ADDRESS_SPACE_MEMORY, cirrus_pci_mmio_map);
      }
-
-     /* ROM BIOS */
-     rom_add_vga(VGABIOS_CIRRUS_FILENAME);
      return 0;
 }
 
 void pci_cirrus_vga_init(PCIBus *bus)
 {
-    pci_create_simple(bus, -1, "Cirrus VGA");
+    pci_create_simple(bus, -1, "cirrus-vga");
 }
 
 static PCIDeviceInfo cirrus_vga_info = {
-    .qdev.name    = "Cirrus VGA",
+    .qdev.name    = "cirrus-vga",
+    .qdev.desc    = "Cirrus CLGD 54xx VGA",
     .qdev.size    = sizeof(PCICirrusVGAState),
     .qdev.vmsd    = &vmstate_pci_cirrus_vga,
     .init         = pci_cirrus_vga_initfn,
+    .romfile      = VGABIOS_CIRRUS_FILENAME,
     .config_write = pci_cirrus_write_config,
 };
 

hw/device-assignment.h

@@ -0,0 +1,117 @@
+/*
+ * Copyright (c) 2007, Neocleus Corporation.
+ * Copyright (c) 2007, Intel Corporation.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms and conditions of the GNU General Public License,
+ * version 2, as published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope it will be useful, but WITHOUT
+ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
+ * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License for
+ * more details.
+ *
+ * You should have received a copy of the GNU General Public License along with
+ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
+ * Place - Suite 330, Boston, MA 02111-1307 USA.
+ *
+ *  Data structures for storing PCI state
+ *
+ *  Adapted to kvm by Qumranet
+ *
+ *  Copyright (c) 2007, Neocleus, Alex Novik (alex@neocleus.com)
+ *  Copyright (c) 2007, Neocleus, Guy Zana (guy@neocleus.com)
+ *  Copyright (C) 2008, Qumranet, Amit Shah (amit.shah@qumranet.com)
+ *  Copyright (C) 2008, Red Hat, Amit Shah (amit.shah@redhat.com)
+ */
+
+#ifndef __DEVICE_ASSIGNMENT_H__
+#define __DEVICE_ASSIGNMENT_H__
+
+#include <sys/mman.h>
+#include "qemu-common.h"
+#include "qemu-queue.h"
+#include "pci.h"
+
+/* From include/linux/pci.h in the kernel sources */
+#define PCI_DEVFN(slot, func)   ((((slot) & 0x1f) << 3) | ((func) & 0x07))
+
+typedef struct PCIHostDevice {
+    int bus;
+    int dev;
+    int func;
+} PCIHostDevice;
+
+typedef struct {
+    int type;           /* Memory or port I/O */
+    int valid;
+    uint32_t base_addr;
+    uint32_t size;    /* size of the region */
+    int resource_fd;
+} PCIRegion;
+
+typedef struct {
+    uint8_t bus, dev, func; /* Bus inside domain, device and function */
+    int irq;                /* IRQ number */
+    uint16_t region_number; /* number of active regions */
+
+    /* Port I/O or MMIO Regions */
+    PCIRegion regions[PCI_NUM_REGIONS];
+    int config_fd;
+} PCIDevRegions;
+
+typedef struct {
+    pcibus_t e_physbase;
+    uint32_t memory_index;
+    union {
+        void *r_virtbase;    /* mmapped access address for memory regions */
+        uint32_t r_baseport; /* the base guest port for I/O regions */
+    } u;
+    int num;            /* our index within v_addrs[] */
+    pcibus_t e_size;    /* emulated size of region in bytes */
+    pcibus_t r_size;    /* real size of region in bytes */
+} AssignedDevRegion;
+
+typedef struct AssignedDevice {
+    PCIDevice dev;
+    PCIHostDevice host;
+    uint32_t use_iommu;
+    int intpin;
+    uint8_t debug_flags;
+    AssignedDevRegion v_addrs[PCI_NUM_REGIONS];
+    PCIDevRegions real_device;
+    int run;
+    int girq;
+    unsigned char h_busnr;
+    unsigned int h_devfn;
+    int irq_requested_type;
+    int bound;
+    struct pci_dev *pdev;
+    struct {
+#define ASSIGNED_DEVICE_CAP_MSI (1 << 0)
+#define ASSIGNED_DEVICE_CAP_MSIX (1 << 1)
+        uint32_t available;
+#define ASSIGNED_DEVICE_MSI_ENABLED (1 << 0)
+#define ASSIGNED_DEVICE_MSIX_ENABLED (1 << 1)
+#define ASSIGNED_DEVICE_MSIX_MASKED (1 << 2)
+        uint32_t state;
+    } cap;
+    int irq_entries_nr;
+    struct kvm_irq_routing_entry *entry;
+    void *msix_table_page;
+    target_phys_addr_t msix_table_addr;
+    int mmio_index;
+    int need_emulate_cmd;
+    QLIST_ENTRY(AssignedDevice) next;
+} AssignedDevice;
+
+QemuOpts *add_assigned_device(const char *arg);
+void add_assigned_devices(PCIBus *bus, const char **devices, int n_devices);
+void assigned_dev_update_irqs(void);
+
+#define MAX_DEV_ASSIGN_CMDLINE 8
+
+extern const char *assigned_devices[MAX_DEV_ASSIGN_CMDLINE];
+extern int assigned_devices_index;
+
+#endif              /* __DEVICE_ASSIGNMENT_H__ */

hw/e1000.c

@@ -1089,7 +1089,6 @@ static int pci_e1000_init(PCIDevice *pci_dev)
 
     pci_config_set_vendor_id(pci_conf, PCI_VENDOR_ID_INTEL);
     pci_config_set_device_id(pci_conf, E1000_DEVID);
-    *(uint16_t *)(pci_conf+0x04) = cpu_to_le16(0x0407);
     *(uint16_t *)(pci_conf+0x06) = cpu_to_le16(0x0010);
     pci_conf[0x08] = 0x03;
     pci_config_set_class(pci_conf, PCI_CLASS_NETWORK_ETHERNET);
@@ -1121,14 +1120,6 @@ static int pci_e1000_init(PCIDevice *pci_dev)
                           d->dev.qdev.info->name, d->dev.qdev.id, d);
 
     qemu_format_nic_info_str(&d->nic->nc, macaddr);
-
-    if (!pci_dev->qdev.hotplugged) {
-        static int loaded = 0;
-        if (!loaded) {
-            rom_add_option("pxe-e1000.bin");
-            loaded = 1;
-        }
-    }
     return 0;
 }
 
@@ -1146,6 +1137,7 @@ static PCIDeviceInfo e1000_info = {
     .qdev.vmsd  = &vmstate_e1000,
     .init       = pci_e1000_init,
     .exit       = pci_e1000_uninit,
+    .romfile    = "pxe-e1000.bin",
     .qdev.props = (Property[]) {
         DEFINE_NIC_PROPERTIES(E1000State, conf),
         DEFINE_PROP_END_OF_LIST(),

hw/extboot.c

@@ -0,0 +1,135 @@
+/*
+ * Extended boot option ROM support.
+ *
+ * Copyright IBM, Corp. 2007
+ *
+ * Authors:
+ *  Anthony Liguori   <aliguori@us.ibm.com>
+ *
+ * This work is licensed under the terms of the GNU GPL, version 2.  See
+ * the COPYING file in the top-level directory.
+ *
+ */
+
+#include "hw.h"
+#include "pc.h"
+#include "isa.h"
+#include "block.h"
+
+/* Extended Boot ROM suport */
+
+union extboot_cmd
+{
+    uint16_t type;
+    struct {
+	uint16_t type;
+	uint16_t cylinders;
+	uint16_t heads;
+	uint16_t sectors;
+	uint64_t nb_sectors;
+    } query_geometry;
+    struct {
+	uint16_t type;
+	uint16_t nb_sectors;
+	uint16_t segment;
+	uint16_t offset;
+	uint64_t sector;
+    } xfer;
+};
+
+static void get_translated_chs(BlockDriverState *bs, int *c, int *h, int *s)
+{
+    bdrv_get_geometry_hint(bs, c, h, s);
+
+    if (*c <= 1024) {
+	*c >>= 0;
+	*h <<= 0;
+    } else if (*c <= 2048) {
+	*c >>= 1;
+	*h <<= 1;
+    } else if (*c <= 4096) {
+	*c >>= 2;
+	*h <<= 2;
+    } else if (*c <= 8192) {
+	*c >>= 3;
+	*h <<= 3;
+    } else {
+	*c >>= 4;
+	*h <<= 4;
+    }
+
+    /* what is the correct algorithm for this?? */
+    if (*h == 256) {
+	*h = 255;
+	*c = *c + 1;
+    }
+}
+
+static uint32_t extboot_read(void *opaque, uint32_t addr)
+{
+    int *pcmd = opaque;
+    return *pcmd;
+}
+
+static void extboot_write_cmd(void *opaque, uint32_t addr, uint32_t value)
+{
+    union extboot_cmd cmd;
+    BlockDriverState *bs = opaque;
+    int cylinders, heads, sectors, err;
+    uint64_t nb_sectors;
+    target_phys_addr_t pa = 0;
+    int blen = 0;
+    void *buf = NULL;
+
+    cpu_physical_memory_read((value & 0xFFFF) << 4, (uint8_t *)&cmd,
+                             sizeof(cmd));
+
+    if (cmd.type == 0x01 || cmd.type == 0x02) {
+	pa = cmd.xfer.segment * 16 + cmd.xfer.offset;
+        blen = cmd.xfer.nb_sectors * 512;
+        buf = qemu_memalign(512, blen);
+    }
+
+    switch (cmd.type) {
+    case 0x00:
+        get_translated_chs(bs, &cylinders, &heads, &sectors);
+	bdrv_get_geometry(bs, &nb_sectors);
+	cmd.query_geometry.cylinders = cylinders;
+	cmd.query_geometry.heads = heads;
+	cmd.query_geometry.sectors = sectors;
+	cmd.query_geometry.nb_sectors = nb_sectors;
+	break;
+    case 0x01:
+	err = bdrv_read(bs, cmd.xfer.sector, buf, cmd.xfer.nb_sectors);
+	if (err)
+	    printf("Read failed\n");
+
+        cpu_physical_memory_write(pa, buf, blen);
+
+	break;
+    case 0x02:
+        cpu_physical_memory_read(pa, buf, blen);
+
+	err = bdrv_write(bs, cmd.xfer.sector, buf, cmd.xfer.nb_sectors);
+	if (err)
+	    printf("Write failed\n");
+
+	break;
+    }
+
+    cpu_physical_memory_write((value & 0xFFFF) << 4, (uint8_t *)&cmd,
+                              sizeof(cmd));
+    if (buf)
+        qemu_free(buf);
+}
+
+void extboot_init(BlockDriverState *bs, int cmd)
+{
+    int *pcmd;
+
+    pcmd = qemu_mallocz(sizeof(int));
+
+    *pcmd = cmd;
+    register_ioport_read(0x404, 1, 1, extboot_read, pcmd);
+    register_ioport_write(0x405, 1, 2, extboot_write_cmd, bs);
+}

hw/fdc.c

@@ -370,9 +370,9 @@ enum {
     FD_CMD_PART_ID = 0x18,
     FD_CMD_SCAN_LOW_OR_EQUAL = 0x19,
     FD_CMD_SCAN_HIGH_OR_EQUAL = 0x1d,
-    FD_CMD_SAVE = 0x2c,
+    FD_CMD_SAVE = 0x2e,
     FD_CMD_OPTION = 0x33,
-    FD_CMD_RESTORE = 0x4c,
+    FD_CMD_RESTORE = 0x4e,
     FD_CMD_DRIVE_SPECIFICATION_COMMAND = 0x8e,
     FD_CMD_RELATIVE_SEEK_OUT = 0x8f,
     FD_CMD_FORMAT_AND_WRITE = 0xcd,
@@ -661,7 +661,7 @@ static int fdc_post_load(void *opaque, int version_id)
 }
 
 static const VMStateDescription vmstate_fdc = {
-    .name = "fdctrl",
+    .name = "fdc",
     .version_id = 2,
     .minimum_version_id = 2,
     .minimum_version_id_old = 2,
@@ -699,31 +699,6 @@ static const VMStateDescription vmstate_fdc = {
     }
 };
 
-static const VMStateDescription vmstate_fdc_isa = {
-    .name = "fdc",
-    .version_id = 2,
-    .minimum_version_id = 2,
-    .minimum_version_id_old = 2,
-    .fields      = (VMStateField []) {
-        /* Controller State */
-        VMSTATE_STRUCT(state, fdctrl_isabus_t, 0, vmstate_fdc, fdctrl_t),
-        VMSTATE_END_OF_LIST()
-    }
-};
-
-static const VMStateDescription vmstate_fdc_sysbus = {
-    .name = "fdc",
-    .version_id = 2,
-    .minimum_version_id = 2,
-    .minimum_version_id_old = 2,
-    .fields      = (VMStateField []) {
-        /* Controller State */
-        VMSTATE_STRUCT(state, fdctrl_sysbus_t, 0, vmstate_fdc, fdctrl_t),
-        VMSTATE_END_OF_LIST()
-    }
-};
-
-
 static void fdctrl_external_reset_sysbus(DeviceState *d)
 {
     fdctrl_sysbus_t *sys = container_of(d, fdctrl_sysbus_t, busdev.qdev);
@@ -960,6 +935,12 @@ static uint32_t fdctrl_read_main_status (fdctrl_t *fdctrl)
     fdctrl->dsr &= ~FD_DSR_PWRDOWN;
     fdctrl->dor |= FD_DOR_nRESET;
 
+    /* Sparc mutation */
+    if (fdctrl->sun4m) {
+        retval |= FD_MSR_DIO;
+        fdctrl_reset_irq(fdctrl);
+    };
+
     FLOPPY_DPRINTF("main status register: 0x%02x\n", retval);
 
     return retval;
@@ -1879,8 +1860,12 @@ fdctrl_t *fdctrl_init_isa(DriveInfo **fds)
     ISADevice *dev;
 
     dev = isa_create("isa-fdc");
-    qdev_prop_set_drive(&dev->qdev, "driveA", fds[0]);
+    if (fds[0]) {
-    qdev_prop_set_drive(&dev->qdev, "driveB", fds[1]);
+        qdev_prop_set_drive(&dev->qdev, "driveA", fds[0]);
+    }
+    if (fds[1]) {
+        qdev_prop_set_drive(&dev->qdev, "driveB", fds[1]);
+    }
     if (qdev_init(&dev->qdev) < 0)
         return NULL;
     return &(DO_UPCAST(fdctrl_isabus_t, busdev, dev)->state);
@@ -1898,8 +1883,12 @@ fdctrl_t *fdctrl_init_sysbus(qemu_irq irq, int dma_chann,
     sys = DO_UPCAST(fdctrl_sysbus_t, busdev.qdev, dev);
     fdctrl = &sys->state;
     fdctrl->dma_chann = dma_chann; /* FIXME */
-    qdev_prop_set_drive(dev, "driveA", fds[0]);
+    if (fds[0]) {
-    qdev_prop_set_drive(dev, "driveB", fds[1]);
+        qdev_prop_set_drive(dev, "driveA", fds[0]);
+    }
+    if (fds[1]) {
+        qdev_prop_set_drive(dev, "driveB", fds[1]);
+    }
     qdev_init_nofail(dev);
     sysbus_connect_irq(&sys->busdev, 0, irq);
     sysbus_mmio_map(&sys->busdev, 0, mmio_base);
@@ -1915,7 +1904,9 @@ fdctrl_t *sun4m_fdctrl_init (qemu_irq irq, target_phys_addr_t io_base,
     fdctrl_t *fdctrl;
 
     dev = qdev_create(NULL, "SUNW,fdtwo");
-    qdev_prop_set_drive(dev, "drive", fds[0]);
+    if (fds[0]) {
+        qdev_prop_set_drive(dev, "drive", fds[0]);
+    }
     qdev_init_nofail(dev);
     sys = DO_UPCAST(fdctrl_sysbus_t, busdev.qdev, dev);
     fdctrl = &sys->state;
@@ -1926,7 +1917,7 @@ fdctrl_t *sun4m_fdctrl_init (qemu_irq irq, target_phys_addr_t io_base,
     return fdctrl;
 }
 
-static int fdctrl_init_common(fdctrl_t *fdctrl)
+static int fdctrl_init_common(fdctrl_t *fdctrl, target_phys_addr_t io_base)
 {
     int i, j;
     static int command_tables_inited = 0;
@@ -1957,6 +1948,7 @@ static int fdctrl_init_common(fdctrl_t *fdctrl)
         DMA_register_channel(fdctrl->dma_chann, &fdctrl_transfer_handler, fdctrl);
     fdctrl_connect_drives(fdctrl);
 
+    vmstate_register(io_base, &vmstate_fdc, fdctrl);
     return 0;
 }
 
@@ -1980,7 +1972,7 @@ static int isabus_fdc_init1(ISADevice *dev)
     isa_init_irq(&isa->busdev, &fdctrl->irq, isairq);
     fdctrl->dma_chann = dma_chann;
 
-    ret = fdctrl_init_common(fdctrl);
+    ret = fdctrl_init_common(fdctrl, iobase);
 
     return ret;
 }
@@ -1998,7 +1990,7 @@ static int sysbus_fdc_init1(SysBusDevice *dev)
     qdev_init_gpio_in(&dev->qdev, fdctrl_handle_tc, 1);
     fdctrl->dma_chann = -1;
 
-    ret = fdctrl_init_common(fdctrl);
+    ret = fdctrl_init_common(fdctrl, io);
 
     return ret;
 }
@@ -2015,7 +2007,7 @@ static int sun4m_fdc_init1(SysBusDevice *dev)
     qdev_init_gpio_in(&dev->qdev, fdctrl_handle_tc, 1);
 
     fdctrl->sun4m = 1;
-    return fdctrl_init_common(fdctrl);
+    return fdctrl_init_common(fdctrl, io);
 }
 
 static ISADeviceInfo isa_fdc_info = {
@@ -2023,7 +2015,6 @@ static ISADeviceInfo isa_fdc_info = {
     .qdev.name  = "isa-fdc",
     .qdev.size  = sizeof(fdctrl_isabus_t),
     .qdev.no_user = 1,
-    .qdev.vmsd  = &vmstate_fdc_isa,
     .qdev.reset = fdctrl_external_reset_isa,
     .qdev.props = (Property[]) {
         DEFINE_PROP_DRIVE("driveA", fdctrl_isabus_t, state.drives[0].dinfo),
@@ -2036,7 +2027,6 @@ static SysBusDeviceInfo sysbus_fdc_info = {
     .init = sysbus_fdc_init1,
     .qdev.name  = "sysbus-fdc",
     .qdev.size  = sizeof(fdctrl_sysbus_t),
-    .qdev.vmsd  = &vmstate_fdc_sysbus,
     .qdev.reset = fdctrl_external_reset_sysbus,
     .qdev.props = (Property[]) {
         DEFINE_PROP_DRIVE("driveA", fdctrl_sysbus_t, state.drives[0].dinfo),
@@ -2049,7 +2039,6 @@ static SysBusDeviceInfo sun4m_fdc_info = {
     .init = sun4m_fdc_init1,
     .qdev.name  = "SUNW,fdtwo",
     .qdev.size  = sizeof(fdctrl_sysbus_t),
-    .qdev.vmsd  = &vmstate_fdc_sysbus,
     .qdev.reset = fdctrl_external_reset_sysbus,
     .qdev.props = (Property[]) {
         DEFINE_PROP_DRIVE("drive", fdctrl_sysbus_t, state.drives[0].dinfo),

hw/fw_cfg.c

@@ -45,11 +45,12 @@ typedef struct _FWCfgEntry {
     FWCfgCallback callback;
 } FWCfgEntry;
 
-typedef struct _FWCfgState {
+struct _FWCfgState {
     FWCfgEntry entries[2][FW_CFG_MAX_ENTRY];
+    FWCfgFiles *files;
     uint16_t cur_entry;
     uint32_t cur_offset;
-} FWCfgState;
+};
 
 static void fw_cfg_write(FWCfgState *s, uint8_t value)
 {
@@ -178,7 +179,7 @@ static int get_uint32_as_uint16(QEMUFile *f, void *pv, size_t size)
 
 static void put_unused(QEMUFile *f, void *pv, size_t size)
 {
-    fprintf(stderr, "uint32_as_uint16 is only used for backward compatibilty.\n");
+    fprintf(stderr, "uint32_as_uint16 is only used for backward compatibility.\n");
     fprintf(stderr, "This functions shouldn't be called.\n");
 }
 
@@ -210,9 +211,8 @@ static const VMStateDescription vmstate_fw_cfg = {
     }
 };
 
-int fw_cfg_add_bytes(void *opaque, uint16_t key, uint8_t *data, uint32_t len)
+int fw_cfg_add_bytes(FWCfgState *s, uint16_t key, uint8_t *data, uint32_t len)
 {
-    FWCfgState *s = opaque;
     int arch = !!(key & FW_CFG_ARCH_LOCAL);
 
     key &= FW_CFG_ENTRY_MASK;
@@ -226,37 +226,36 @@ int fw_cfg_add_bytes(void *opaque, uint16_t key, uint8_t *data, uint32_t len)
     return 1;
 }
 
-int fw_cfg_add_i16(void *opaque, uint16_t key, uint16_t value)
+int fw_cfg_add_i16(FWCfgState *s, uint16_t key, uint16_t value)
 {
     uint16_t *copy;
 
     copy = qemu_malloc(sizeof(value));
     *copy = cpu_to_le16(value);
-    return fw_cfg_add_bytes(opaque, key, (uint8_t *)copy, sizeof(value));
+    return fw_cfg_add_bytes(s, key, (uint8_t *)copy, sizeof(value));
 }
 
-int fw_cfg_add_i32(void *opaque, uint16_t key, uint32_t value)
+int fw_cfg_add_i32(FWCfgState *s, uint16_t key, uint32_t value)
 {
     uint32_t *copy;
 
     copy = qemu_malloc(sizeof(value));
     *copy = cpu_to_le32(value);
-    return fw_cfg_add_bytes(opaque, key, (uint8_t *)copy, sizeof(value));
+    return fw_cfg_add_bytes(s, key, (uint8_t *)copy, sizeof(value));
 }
 
-int fw_cfg_add_i64(void *opaque, uint16_t key, uint64_t value)
+int fw_cfg_add_i64(FWCfgState *s, uint16_t key, uint64_t value)
 {
     uint64_t *copy;
 
     copy = qemu_malloc(sizeof(value));
     *copy = cpu_to_le64(value);
-    return fw_cfg_add_bytes(opaque, key, (uint8_t *)copy, sizeof(value));
+    return fw_cfg_add_bytes(s, key, (uint8_t *)copy, sizeof(value));
 }
 
-int fw_cfg_add_callback(void *opaque, uint16_t key, FWCfgCallback callback,
+int fw_cfg_add_callback(FWCfgState *s, uint16_t key, FWCfgCallback callback,
                         void *callback_opaque, uint8_t *data, size_t len)
 {
-    FWCfgState *s = opaque;
     int arch = !!(key & FW_CFG_ARCH_LOCAL);
 
     if (!(key & FW_CFG_WRITE_CHANNEL))
@@ -275,8 +274,54 @@ int fw_cfg_add_callback(void *opaque, uint16_t key, FWCfgCallback callback,
     return 1;
 }
 
-void *fw_cfg_init(uint32_t ctl_port, uint32_t data_port,
+int fw_cfg_add_file(FWCfgState *s,  const char *dir, const char *filename,
-		target_phys_addr_t ctl_addr, target_phys_addr_t data_addr)
+                    uint8_t *data, uint32_t len)
+{
+    const char *basename;
+    int i, index;
+
+    if (!s->files) {
+        int dsize = sizeof(uint32_t) + sizeof(FWCfgFile) * FW_CFG_FILE_SLOTS;
+        s->files = qemu_mallocz(dsize);
+        fw_cfg_add_bytes(s, FW_CFG_FILE_DIR, (uint8_t*)s->files, dsize);
+    }
+
+    index = be32_to_cpu(s->files->count);
+    if (index == FW_CFG_FILE_SLOTS) {
+        fprintf(stderr, "fw_cfg: out of file slots\n");
+        return 0;
+    }
+
+    fw_cfg_add_bytes(s, FW_CFG_FILE_FIRST + index, data, len);
+
+    basename = strrchr(filename, '/');
+    if (basename) {
+        basename++;
+    } else {
+        basename = filename;
+    }
+
+    snprintf(s->files->f[index].name, sizeof(s->files->f[index].name),
+             "%s/%s", dir, basename);
+    for (i = 0; i < index; i++) {
+        if (strcmp(s->files->f[index].name, s->files->f[i].name) == 0) {
+            FW_CFG_DPRINTF("%s: skip duplicate: %s\n", __FUNCTION__,
+                           s->files->f[index].name);
+            return 1;
+        }
+    }
+
+    s->files->f[index].size   = cpu_to_be32(len);
+    s->files->f[index].select = cpu_to_be16(FW_CFG_FILE_FIRST + index);
+    FW_CFG_DPRINTF("%s: #%d: %s (%d bytes)\n", __FUNCTION__,
+                   index, s->files->f[index].name, len);
+
+    s->files->count = cpu_to_be32(index+1);
+    return 1;
+}
+
+FWCfgState *fw_cfg_init(uint32_t ctl_port, uint32_t data_port,
+                        target_phys_addr_t ctl_addr, target_phys_addr_t data_addr)
 {
     FWCfgState *s;
     int io_ctl_memory, io_data_memory;

hw/fw_cfg.h

@@ -26,7 +26,11 @@
 #define FW_CFG_SETUP_ADDR       0x16
 #define FW_CFG_SETUP_SIZE       0x17
 #define FW_CFG_SETUP_DATA       0x18
-#define FW_CFG_MAX_ENTRY        0x19
+#define FW_CFG_FILE_DIR         0x19
+
+#define FW_CFG_FILE_FIRST       0x20
+#define FW_CFG_FILE_SLOTS       0x10
+#define FW_CFG_MAX_ENTRY        (FW_CFG_FILE_FIRST+FW_CFG_FILE_SLOTS)
 
 #define FW_CFG_WRITE_CHANNEL    0x4000
 #define FW_CFG_ARCH_LOCAL       0x8000
@@ -35,16 +39,31 @@
 #define FW_CFG_INVALID          0xffff
 
 #ifndef NO_QEMU_PROTOS
+typedef struct FWCfgFile {
+    uint32_t  size;        /* file size */
+    uint16_t  select;      /* write this to 0x510 to read it */
+    uint16_t  reserved;
+    char      name[56];
+} FWCfgFile;
+
+typedef struct FWCfgFiles {
+    uint32_t  count;
+    FWCfgFile f[];
+} FWCfgFiles;
+
 typedef void (*FWCfgCallback)(void *opaque, uint8_t *data);
 
-int fw_cfg_add_bytes(void *opaque, uint16_t key, uint8_t *data, uint32_t len);
+typedef struct _FWCfgState FWCfgState;
-int fw_cfg_add_i16(void *opaque, uint16_t key, uint16_t value);
+int fw_cfg_add_bytes(FWCfgState *s, uint16_t key, uint8_t *data, uint32_t len);
-int fw_cfg_add_i32(void *opaque, uint16_t key, uint32_t value);
+int fw_cfg_add_i16(FWCfgState *s, uint16_t key, uint16_t value);
-int fw_cfg_add_i64(void *opaque, uint16_t key, uint64_t value);
+int fw_cfg_add_i32(FWCfgState *s, uint16_t key, uint32_t value);
-int fw_cfg_add_callback(void *opaque, uint16_t key, FWCfgCallback callback,
+int fw_cfg_add_i64(FWCfgState *s, uint16_t key, uint64_t value);
+int fw_cfg_add_callback(FWCfgState *s, uint16_t key, FWCfgCallback callback,
                         void *callback_opaque, uint8_t *data, size_t len);
-void *fw_cfg_init(uint32_t ctl_port, uint32_t data_port,
+int fw_cfg_add_file(FWCfgState *s, const char *dir, const char *filename,
-		target_phys_addr_t crl_addr, target_phys_addr_t data_addr);
+                    uint8_t *data, uint32_t len);
+FWCfgState *fw_cfg_init(uint32_t ctl_port, uint32_t data_port,
+                        target_phys_addr_t crl_addr, target_phys_addr_t data_addr);
 
 #endif /* NO_QEMU_PROTOS */
 

hw/grackle_pci.c

@@ -178,7 +178,7 @@ static PCIDeviceInfo grackle_pci_host_info = {
 };
 
 static PCIDeviceInfo dec_21154_pci_host_info = {
-    .qdev.name = "DEC 21154",
+    .qdev.name = "dec-21154",
     .qdev.size = sizeof(PCIDevice),
     .init      = dec_21154_pci_host_init,
 };
@@ -188,7 +188,7 @@ static void grackle_register_devices(void)
     sysbus_register_dev("grackle", sizeof(GrackleState),
                         pci_grackle_init_device);
     pci_qdev_register(&grackle_pci_host_info);
-    sysbus_register_dev("DEC 21154", sizeof(GrackleState),
+    sysbus_register_dev("dec-21154", sizeof(GrackleState),
                         pci_dec_21154_init_device);
     pci_qdev_register(&dec_21154_pci_host_info);
 }

hw/hpet.c

@@ -170,6 +170,11 @@ static int hpet_post_load(void *opaque, int version_id)
 
     /* Recalculate the offset between the main counter and guest time */
     s->hpet_offset = ticks_to_ns(s->hpet_counter) - qemu_get_clock(vm_clock);
+
+    if (hpet_in_legacy_mode()) {
+        hpet_disable_pit();
+    }
+
     return 0;
 }
 
@@ -473,9 +478,11 @@ static void hpet_ram_writel(void *opaque, target_phys_addr_t addr,
                 }
                 /* i8254 and RTC are disabled when HPET is in legacy mode */
                 if (activating_bit(old_val, new_val, HPET_CFG_LEGACY)) {
-                    hpet_pit_disable();
+                    hpet_disable_pit();
+                    dprintf("qemu: hpet disabled pit\n");
                 } else if (deactivating_bit(old_val, new_val, HPET_CFG_LEGACY)) {
-                    hpet_pit_enable();
+                    hpet_enable_pit();
+                    dprintf("qemu: hpet enabled pit\n");
                 }
                 break;
             case HPET_CFG + 4:
@@ -559,7 +566,7 @@ static void hpet_reset(void *opaque) {
          * hpet_reset is called due to system reset. At this point control must
          * be returned to pit until SW reenables hpet.
          */
-        hpet_pit_enable();
+        hpet_enable_pit();
     count = 1;
 }
 

hw/hw.h

@@ -342,6 +342,10 @@ extern const VMStateInfo vmstate_info_uint16;
 extern const VMStateInfo vmstate_info_uint32;
 extern const VMStateInfo vmstate_info_uint64;
 
+#ifdef __linux__
+extern const VMStateInfo vmstate_info_u64;
+#endif
+
 extern const VMStateInfo vmstate_info_timer;
 extern const VMStateInfo vmstate_info_ptimer;
 extern const VMStateInfo vmstate_info_buffer;
@@ -622,6 +626,15 @@ extern const VMStateDescription vmstate_i2c_slave;
 #define VMSTATE_UINT64(_f, _s)                                        \
     VMSTATE_UINT64_V(_f, _s, 0)
 
+/* This is needed because on linux __u64 is unsigned long long
+   and on glibc uint64_t is unsigned long on 64 bits */
+#ifdef __linux__
+#define VMSTATE_U64_V(_f, _s, _v)                                     \
+    VMSTATE_SINGLE(_f, _s, _v, vmstate_info_u64, __u64)
+#define VMSTATE_U64(_f, _s)                                           \
+    VMSTATE_U64_V(_f, _s, 0)
+#endif
+
 #define VMSTATE_UINT8_EQUAL(_f, _s)                                   \
     VMSTATE_SINGLE(_f, _s, 0, vmstate_info_uint8_equal, uint8_t)
 

hw/i8254-kvm.c

@@ -0,0 +1,122 @@
+/*
+ * QEMU 8253/8254 interval timer emulation
+ *
+ * Copyright (c) 2003-2004 Fabrice Bellard
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+#include "hw.h"
+#include "pc.h"
+#include "isa.h"
+#include "qemu-timer.h"
+#include "i8254.h"
+#include "qemu-kvm.h"
+
+extern VMStateDescription vmstate_pit;
+
+static PITState pit_state;
+
+static void kvm_pit_pre_save(void *opaque)
+{
+    PITState *s = (void *)opaque;
+    struct kvm_pit_state2 pit2;
+    struct kvm_pit_channel_state *c;
+    struct PITChannelState *sc;
+    int i;
+
+    if(qemu_kvm_has_pit_state2()) {
+        kvm_get_pit2(kvm_context, &pit2);
+        s->flags = pit2.flags;
+    } else {
+        /* pit2 is superset of pit struct so just cast it and use it */
+        kvm_get_pit(kvm_context, (struct kvm_pit_state *)&pit2);
+    }
+    for (i = 0; i < 3; i++) {
+	c = &pit2.channels[i];
+	sc = &s->channels[i];
+	sc->count = c->count;
+	sc->latched_count = c->latched_count;
+	sc->count_latched = c->count_latched;
+	sc->status_latched = c->status_latched;
+	sc->status = c->status;
+	sc->read_state = c->read_state;
+	sc->write_state = c->write_state;
+	sc->write_latch = c->write_latch;
+	sc->rw_mode = c->rw_mode;
+	sc->mode = c->mode;
+	sc->bcd = c->bcd;
+	sc->gate = c->gate;
+	sc->count_load_time = c->count_load_time;
+    }
+}
+
+static int kvm_pit_post_load(void *opaque, int version_id)
+{
+    PITState *s = opaque;
+    struct kvm_pit_state2 pit2;
+    struct kvm_pit_channel_state *c;
+    struct PITChannelState *sc;
+    int i;
+
+    pit2.flags = s->flags;
+    for (i = 0; i < 3; i++) {
+	c = &pit2.channels[i];
+	sc = &s->channels[i];
+	c->count = sc->count;
+	c->latched_count = sc->latched_count;
+	c->count_latched = sc->count_latched;
+	c->status_latched = sc->status_latched;
+	c->status = sc->status;
+	c->read_state = sc->read_state;
+	c->write_state = sc->write_state;
+	c->write_latch = sc->write_latch;
+	c->rw_mode = sc->rw_mode;
+	c->mode = sc->mode;
+	c->bcd = sc->bcd;
+	c->gate = sc->gate;
+	c->count_load_time = sc->count_load_time;
+    }
+
+    if(qemu_kvm_has_pit_state2()) {
+        kvm_set_pit2(kvm_context, &pit2);
+    } else {
+        kvm_set_pit(kvm_context, (struct kvm_pit_state *)&pit2);
+    }
+    return 0;
+}
+
+static void dummy_timer(void *opaque)
+{
+}
+
+PITState *kvm_pit_init(int base, qemu_irq irq)
+{
+    PITState *pit = &pit_state;
+    PITChannelState *s;
+
+    s = &pit->channels[0];
+    s->irq_timer = qemu_new_timer(vm_clock, dummy_timer, s);
+    vmstate_pit.pre_save = kvm_pit_pre_save;
+    vmstate_pit.post_load = kvm_pit_post_load;
+    vmstate_register(base, &vmstate_pit, pit);
+    qemu_register_reset(pit_reset, pit);
+    pit_reset(pit);
+
+    return pit;
+}

hw/i8254.c

@@ -25,38 +25,11 @@
 #include "pc.h"
 #include "isa.h"
 #include "qemu-timer.h"
+#include "qemu-kvm.h"
+#include "i8254.h"
 
 //#define DEBUG_PIT
 
-#define RW_STATE_LSB 1
-#define RW_STATE_MSB 2
-#define RW_STATE_WORD0 3
-#define RW_STATE_WORD1 4
-
-typedef struct PITChannelState {
-    int count; /* can be 65536 */
-    uint16_t latched_count;
-    uint8_t count_latched;
-    uint8_t status_latched;
-    uint8_t status;
-    uint8_t read_state;
-    uint8_t write_state;
-    uint8_t write_latch;
-    uint8_t rw_mode;
-    uint8_t mode;
-    uint8_t bcd; /* not supported */
-    uint8_t gate; /* timer start */
-    int64_t count_load_time;
-    /* irq handling */
-    int64_t next_transition_time;
-    QEMUTimer *irq_timer;
-    qemu_irq irq;
-} PITChannelState;
-
-struct PITState {
-    PITChannelState channels[3];
-};
-
 static PITState pit_state;
 
 static void pit_irq_timer_update(PITChannelState *s, int64_t current_time);
@@ -228,13 +201,18 @@ int pit_get_mode(PITState *pit, int channel)
     return s->mode;
 }
 
-static inline void pit_load_count(PITChannelState *s, int val)
+static inline void pit_load_count(PITState *s, int val, int chan)
 {
     if (val == 0)
         val = 0x10000;
-    s->count_load_time = qemu_get_clock(vm_clock);
+    s->channels[chan].count_load_time = qemu_get_clock(vm_clock);
-    s->count = val;
+    s->channels[chan].count = val;
-    pit_irq_timer_update(s, s->count_load_time);
+#ifdef TARGET_I386
+    if (chan == 0 && pit_state.flags & PIT_FLAGS_HPET_LEGACY) {
+        return;
+    }
+#endif
+    pit_irq_timer_update(&s->channels[chan], s->channels[chan].count_load_time);
 }
 
 /* if already latched, do not latch again */
@@ -294,17 +272,17 @@ static void pit_ioport_write(void *opaque, uint32_t addr, uint32_t val)
         switch(s->write_state) {
         default:
         case RW_STATE_LSB:
-            pit_load_count(s, val);
+            pit_load_count(pit, val, addr);
             break;
         case RW_STATE_MSB:
-            pit_load_count(s, val << 8);
+            pit_load_count(pit, val << 8, addr);
             break;
         case RW_STATE_WORD0:
             s->write_latch = val;
             s->write_state = RW_STATE_WORD1;
             break;
         case RW_STATE_WORD1:
-            pit_load_count(s, s->write_latch | (val << 8));
+            pit_load_count(pit, s->write_latch | (val << 8), addr);
             s->write_state = RW_STATE_WORD0;
             break;
         }
@@ -364,6 +342,11 @@ static uint32_t pit_ioport_read(void *opaque, uint32_t addr)
     return ret;
 }
 
+/* global counters for time-drift fix */
+int64_t timer_acks=0, timer_interrupts=0, timer_ints_to_push=0;
+
+extern int time_drift_fix;
+
 static void pit_irq_timer_update(PITChannelState *s, int64_t current_time)
 {
     int64_t expire_time;
@@ -374,16 +357,35 @@ static void pit_irq_timer_update(PITChannelState *s, int64_t current_time)
     expire_time = pit_get_next_transition_time(s, current_time);
     irq_level = pit_get_out1(s, current_time);
     qemu_set_irq(s->irq, irq_level);
+    if (time_drift_fix && irq_level==1) {
+        /* FIXME: fine tune timer_max_fix (max fix per tick). 
+         *        Should it be 1 (double time), 2 , 4, 10 ? 
+         *        Currently setting it to 5% of PIT-ticks-per-second (per PIT-tick)
+         */
+        const long pit_ticks_per_sec = (s->count>0) ? (PIT_FREQ/s->count) : 0;
+        const long timer_max_fix = pit_ticks_per_sec/20;
+        const long delta = timer_interrupts - timer_acks;
+        const long max_delta = pit_ticks_per_sec * 60; /* one minute */
+        if ((delta >  max_delta) && (pit_ticks_per_sec > 0)) {
+            printf("time drift is too long, %ld seconds were lost\n", delta/pit_ticks_per_sec);
+            timer_acks = timer_interrupts;
+            timer_ints_to_push = 0;
+        } else if (delta > 0) {
+            timer_ints_to_push = MIN(delta, timer_max_fix);
+        }
+        timer_interrupts++;
+    }
 #ifdef DEBUG_PIT
     printf("irq_level=%d next_delay=%f\n",
            irq_level,
            (double)(expire_time - current_time) / get_ticks_per_sec());
 #endif
     s->next_transition_time = expire_time;
-    if (expire_time != -1)
+    if (expire_time != -1) {
         qemu_mod_timer(s->irq_timer, expire_time);
-    else
+    } else {
         qemu_del_timer(s->irq_timer);
+    }
 }
 
 static void pit_irq_timer(void *opaque)
@@ -423,9 +425,10 @@ static int pit_load_old(QEMUFile *f, void *opaque, int version_id)
     PITChannelState *s;
     int i;
 
-    if (version_id != 1)
+    if (version_id != PIT_SAVEVM_VERSION)
         return -EINVAL;
 
+    pit->flags = qemu_get_be32(f);
     for(i = 0; i < 3; i++) {
         s = &pit->channels[i];
         s->count=qemu_get_be32(f);
@@ -446,57 +449,85 @@ static int pit_load_old(QEMUFile *f, void *opaque, int version_id)
             qemu_get_timer(f, s->irq_timer);
         }
     }
+
     return 0;
 }
 
-static const VMStateDescription vmstate_pit = {
+VMStateDescription vmstate_pit = {
     .name = "i8254",
     .version_id = 2,
     .minimum_version_id = 2,
     .minimum_version_id_old = 1,
     .load_state_old = pit_load_old,
     .fields      = (VMStateField []) {
+        VMSTATE_UINT32(flags, PITState),
         VMSTATE_STRUCT_ARRAY(channels, PITState, 3, 2, vmstate_pit_channel, PITChannelState),
         VMSTATE_TIMER(channels[0].irq_timer, PITState),
         VMSTATE_END_OF_LIST()
     }
 };
 
-static void pit_reset(void *opaque)
+void pit_reset(void *opaque)
 {
     PITState *pit = opaque;
     PITChannelState *s;
     int i;
 
+#ifdef TARGET_I386
+    pit->flags &= ~PIT_FLAGS_HPET_LEGACY;
+#endif
     for(i = 0;i < 3; i++) {
         s = &pit->channels[i];
         s->mode = 3;
         s->gate = (i != 2);
-        pit_load_count(s, 0);
+        pit_load_count(pit, 0, i);
     }
 }
 
+#ifdef TARGET_I386
 /* When HPET is operating in legacy mode, i8254 timer0 is disabled */
-void hpet_pit_disable(void) {
+
-    PITChannelState *s;
+void hpet_disable_pit(void)
-    s = &pit_state.channels[0];
+{
-    if (s->irq_timer)
+    PITChannelState *s = &pit_state.channels[0];
-        qemu_del_timer(s->irq_timer);
+
+    if (kvm_enabled() && qemu_kvm_pit_in_kernel()) {
+        if (qemu_kvm_has_pit_state2()) {
+            kvm_hpet_disable_kpit();
+        } else {
+             fprintf(stderr, "%s: kvm does not support pit_state2!\n", __FUNCTION__);
+             exit(1);
+        }
+    } else {
+        pit_state.flags |= PIT_FLAGS_HPET_LEGACY;
+        if (s->irq_timer) {
+            qemu_del_timer(s->irq_timer);
+        }
+    }
 }
 
 /* When HPET is reset or leaving legacy mode, it must reenable i8254
  * timer 0
  */
 
-void hpet_pit_enable(void)
+void hpet_enable_pit(void)
 {
     PITState *pit = &pit_state;
-    PITChannelState *s;
+    PITChannelState *s = &pit->channels[0];
-    s = &pit->channels[0];
+
-    s->mode = 3;
+    if (kvm_enabled() && qemu_kvm_pit_in_kernel()) {
-    s->gate = 1;
+        if (qemu_kvm_has_pit_state2()) {
-    pit_load_count(s, 0);
+            kvm_hpet_enable_kpit();
+        } else {
+             fprintf(stderr, "%s: kvm does not support pit_state2!\n", __FUNCTION__);
+             exit(1);
+        }
+    } else {
+        pit_state.flags &= ~PIT_FLAGS_HPET_LEGACY;
+        pit_load_count(pit, s->count, 0);
+    }
 }
+#endif
 
 PITState *pit_init(int base, qemu_irq irq)
 {

hw/i8254.h

@@ -0,0 +1,69 @@
+/*
+ * QEMU 8253/8254 interval timer emulation
+ *
+ * Copyright (c) 2003-2004 Fabrice Bellard
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#ifndef QEMU_I8254_H
+#define QEMU_I8254_H
+
+#define PIT_SAVEVM_NAME "i8254"
+#define PIT_SAVEVM_VERSION 2
+
+#define RW_STATE_LSB 1
+#define RW_STATE_MSB 2
+#define RW_STATE_WORD0 3
+#define RW_STATE_WORD1 4
+
+#define PIT_FLAGS_HPET_LEGACY  1
+
+typedef struct PITChannelState {
+    int count; /* can be 65536 */
+    uint16_t latched_count;
+    uint8_t count_latched;
+    uint8_t status_latched;
+    uint8_t status;
+    uint8_t read_state;
+    uint8_t write_state;
+    uint8_t write_latch;
+    uint8_t rw_mode;
+    uint8_t mode;
+    uint8_t bcd; /* not supported */
+    uint8_t gate; /* timer start */
+    int64_t count_load_time;
+    /* irq handling */
+    int64_t next_transition_time;
+    QEMUTimer *irq_timer;
+    qemu_irq irq;
+} PITChannelState;
+
+struct PITState {
+    PITChannelState channels[3];
+    uint32_t flags;
+};
+
+void pit_save(QEMUFile *f, void *opaque);
+
+int pit_load(QEMUFile *f, void *opaque, int version_id);
+
+void pit_reset(void *opaque);
+
+#endif

hw/i8259.c

@@ -27,6 +27,8 @@
 #include "monitor.h"
 #include "qemu-timer.h"
 
+#include "qemu-kvm.h"
+
 /* debug PIC */
 //#define DEBUG_PIC
 
@@ -181,7 +183,6 @@ int64_t irq_time[16];
 static void i8259_set_irq(void *opaque, int irq, int level)
 {
     PicState2 *s = opaque;
-
 #if defined(DEBUG_PIC) || defined(DEBUG_IRQ_COUNT)
     if (level != irq_level[irq]) {
 #if defined(DEBUG_PIC)
@@ -212,18 +213,35 @@ static inline void pic_intack(PicState *s, int irq)
     } else {
         s->isr |= (1 << irq);
     }
+
     /* We don't clear a level sensitive interrupt here */
     if (!(s->elcr & (1 << irq)))
         s->irr &= ~(1 << irq);
+
 }
 
+extern int time_drift_fix;
+
 int pic_read_irq(PicState2 *s)
 {
     int irq, irq2, intno;
 
     irq = pic_get_irq(&s->pics[0]);
     if (irq >= 0) {
+
         pic_intack(&s->pics[0], irq);
+#ifndef TARGET_IA64
+	if (time_drift_fix && irq == 0) {
+	    extern int64_t timer_acks, timer_ints_to_push;
+	    timer_acks++;
+	    if (timer_ints_to_push > 0) {
+		timer_ints_to_push--;
+                /* simulate an edge irq0, like the one generated by i8254 */
+                pic_set_irq1(&s->pics[0], 0, 0);
+                pic_set_irq1(&s->pics[0], 0, 1);
+	    }
+	}
+#endif
         if (irq == 2) {
             irq2 = pic_get_irq(&s->pics[1]);
             if (irq2 >= 0) {
@@ -446,9 +464,33 @@ static uint32_t elcr_ioport_read(void *opaque, uint32_t addr1)
     return s->elcr;
 }
 
+static void kvm_kernel_pic_save_to_user(PicState *s);
+static int kvm_kernel_pic_load_from_user(PicState *s);
+
+static void pic_pre_save(void *opaque)
+{
+    PicState *s = opaque;
+
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_pic_save_to_user(s);
+    }
+}
+
+static int pic_post_load(void *opaque, int version_id)
+{
+    PicState *s = opaque;
+
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_pic_load_from_user(s);
+    }
+    return 0;
+}
+
 static const VMStateDescription vmstate_pic = {
     .name = "i8259",
     .version_id = 1,
+    .pre_save = pic_pre_save,
+    .post_load = pic_post_load,
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
     .fields      = (VMStateField []) {
@@ -535,3 +577,103 @@ qemu_irq *i8259_init(qemu_irq parent_irq)
     isa_pic = s;
     return qemu_allocate_irqs(i8259_set_irq, s, 16);
 }
+
+static void kvm_kernel_pic_save_to_user(PicState *s)
+{
+#ifdef KVM_CAP_IRQCHIP
+    struct kvm_irqchip chip;
+    struct kvm_pic_state *kpic;
+
+    chip.chip_id = (&s->pics_state->pics[0] == s) ?
+                   KVM_IRQCHIP_PIC_MASTER :
+                   KVM_IRQCHIP_PIC_SLAVE;
+    kvm_get_irqchip(kvm_context, &chip);
+    kpic = &chip.chip.pic;
+
+    s->last_irr = kpic->last_irr;
+    s->irr = kpic->irr;
+    s->imr = kpic->imr;
+    s->isr = kpic->isr;
+    s->priority_add = kpic->priority_add;
+    s->irq_base = kpic->irq_base;
+    s->read_reg_select = kpic->read_reg_select;
+    s->poll = kpic->poll;
+    s->special_mask = kpic->special_mask;
+    s->init_state = kpic->init_state;
+    s->auto_eoi = kpic->auto_eoi;
+    s->rotate_on_auto_eoi = kpic->rotate_on_auto_eoi;
+    s->special_fully_nested_mode = kpic->special_fully_nested_mode;
+    s->init4 = kpic->init4;
+    s->elcr = kpic->elcr;
+    s->elcr_mask = kpic->elcr_mask;
+#endif
+}
+
+static int kvm_kernel_pic_load_from_user(PicState *s)
+{
+#ifdef KVM_CAP_IRQCHIP
+    struct kvm_irqchip chip;
+    struct kvm_pic_state *kpic;
+
+    chip.chip_id = (&s->pics_state->pics[0] == s) ?
+                   KVM_IRQCHIP_PIC_MASTER :
+                   KVM_IRQCHIP_PIC_SLAVE;
+    kpic = &chip.chip.pic;
+
+    kpic->last_irr = s->last_irr;
+    kpic->irr = s->irr;
+    kpic->imr = s->imr;
+    kpic->isr = s->isr;
+    kpic->priority_add = s->priority_add;
+    kpic->irq_base = s->irq_base;
+    kpic->read_reg_select = s->read_reg_select;
+    kpic->poll = s->poll;
+    kpic->special_mask = s->special_mask;
+    kpic->init_state = s->init_state;
+    kpic->auto_eoi = s->auto_eoi;
+    kpic->rotate_on_auto_eoi = s->rotate_on_auto_eoi;
+    kpic->special_fully_nested_mode = s->special_fully_nested_mode;
+    kpic->init4 = s->init4;
+    kpic->elcr = s->elcr;
+    kpic->elcr_mask = s->elcr_mask;
+
+    kvm_set_irqchip(kvm_context, &chip);
+#endif
+    return 0;
+}
+
+#ifdef KVM_CAP_IRQCHIP
+static void kvm_i8259_set_irq(void *opaque, int irq, int level)
+{
+    int pic_ret;
+    if (kvm_set_irq(irq, level, &pic_ret)) {
+        if (pic_ret != 0)
+            apic_set_irq_delivered();
+        return;
+    }
+}
+
+static void kvm_pic_init1(int io_addr, PicState *s)
+{
+    vmstate_register(io_addr, &vmstate_pic, s);
+    qemu_register_reset(pic_reset, s);
+}
+
+qemu_irq *kvm_i8259_init(qemu_irq parent_irq)
+{
+    PicState2 *s;
+
+    s = qemu_mallocz(sizeof(PicState2));
+
+    kvm_pic_init1(0x20, &s->pics[0]);
+    kvm_pic_init1(0xa0, &s->pics[1]);
+    s->parent_irq = parent_irq;
+    s->pics[0].pics_state = s;
+    s->pics[1].pics_state = s;
+    isa_pic = s;
+    return qemu_allocate_irqs(kvm_i8259_set_irq, s, 24);
+}
+#endif
+
+
+

hw/ide/cmd646.c

@@ -70,11 +70,7 @@ static void ide_map(PCIDevice *pci_dev, int region_num,
 
 static PCIIDEState *pci_from_bm(BMDMAState *bm)
 {
-    if (bm->unit == 0) {
+    return bm->pci_dev;
-        return container_of(bm, PCIIDEState, bmdma[0]);
-    } else {
-        return container_of(bm, PCIIDEState, bmdma[1]);
-    }
 }
 
 static uint32_t bmdma_readb(void *opaque, uint32_t addr)
@@ -145,6 +141,7 @@ static void bmdma_map(PCIDevice *pci_dev, int region_num,
         BMDMAState *bm = &d->bmdma[i];
         d->bus[i].bmdma = bm;
         bm->bus = d->bus+i;
+        bm->pci_dev = d;
         qemu_add_vm_change_state_handler(ide_dma_restart_cb, bm);
 
         register_ioport_write(addr, 1, 1, bmdma_cmd_writeb, bm);
@@ -245,7 +242,7 @@ void pci_cmd646_ide_init(PCIBus *bus, DriveInfo **hd_table,
 {
     PCIDevice *dev;
 
-    dev = pci_create(bus, -1, "CMD646 IDE");
+    dev = pci_create(bus, -1, "cmd646-ide");
     qdev_prop_set_uint32(&dev->qdev, "secondary", secondary_ide_enabled);
     qdev_init_nofail(&dev->qdev);
 
@@ -254,7 +251,7 @@ void pci_cmd646_ide_init(PCIBus *bus, DriveInfo **hd_table,
 
 static PCIDeviceInfo cmd646_ide_info[] = {
     {
-        .qdev.name    = "CMD646 IDE",
+        .qdev.name    = "cmd646-ide",
         .qdev.size    = sizeof(PCIIDEState),
         .init         = pci_cmd646_ide_initfn,
         .qdev.props   = (Property[]) {

hw/ide/core.c

@@ -115,7 +115,7 @@ static void ide_identify(IDEState *s)
     put_le16(p + 20, 3); /* XXX: retired, remove ? */
     put_le16(p + 21, 512); /* cache size in sectors */
     put_le16(p + 22, 4); /* ecc bytes */
-    padstr((char *)(p + 23), QEMU_VERSION, 8); /* firmware version */
+    padstr((char *)(p + 23), s->version, 8); /* firmware version */
     padstr((char *)(p + 27), "QEMU HARDDISK", 40); /* model */
 #if MAX_MULT_SECTORS > 1
     put_le16(p + 47, 0x8000 | MAX_MULT_SECTORS);
@@ -186,7 +186,7 @@ static void ide_atapi_identify(IDEState *s)
     put_le16(p + 20, 3); /* buffer type */
     put_le16(p + 21, 512); /* cache size in sectors */
     put_le16(p + 22, 4); /* ecc bytes */
-    padstr((char *)(p + 23), QEMU_VERSION, 8); /* firmware version */
+    padstr((char *)(p + 23), s->version, 8); /* firmware version */
     padstr((char *)(p + 27), "QEMU DVD-ROM", 40); /* model */
     put_le16(p + 48, 1); /* dword I/O (XXX: should not be set on CDROM) */
 #ifdef USE_DMA_CDROM
@@ -238,7 +238,7 @@ static void ide_cfata_identify(IDEState *s)
     put_le16(p + 8, s->nb_sectors);		/* Sectors per card */
     padstr((char *)(p + 10), s->drive_serial_str, 20); /* serial number */
     put_le16(p + 22, 0x0004);			/* ECC bytes */
-    padstr((char *) (p + 23), QEMU_VERSION, 8);	/* Firmware Revision */
+    padstr((char *) (p + 23), s->version, 8);	/* Firmware Revision */
     padstr((char *) (p + 27), "QEMU MICRODRIVE", 40);/* Model number */
 #if MAX_MULT_SECTORS > 1
     put_le16(p + 47, 0x8000 | MAX_MULT_SECTORS);
@@ -1591,7 +1591,7 @@ static void ide_atapi_cmd(IDEState *s)
         buf[7] = 0; /* reserved */
         padstr8(buf + 8, 8, "QEMU");
         padstr8(buf + 16, 16, "QEMU DVD-ROM");
-        padstr8(buf + 32, 4, QEMU_VERSION);
+        padstr8(buf + 32, 4, s->version);
         ide_atapi_cmd_reply(s, 36, max_len);
         break;
     case GPCMD_GET_CONFIGURATION:
@@ -2590,7 +2590,7 @@ void ide_bus_reset(IDEBus *bus)
     ide_clear_hob(bus);
 }
 
-void ide_init_drive(IDEState *s, DriveInfo *dinfo)
+void ide_init_drive(IDEState *s, DriveInfo *dinfo, const char *version)
 {
     int cylinders, heads, secs;
     uint64_t nb_sectors;
@@ -2619,6 +2619,11 @@ void ide_init_drive(IDEState *s, DriveInfo *dinfo)
     if (strlen(s->drive_serial_str) == 0)
         snprintf(s->drive_serial_str, sizeof(s->drive_serial_str),
                  "QM%05d", s->drive_serial);
+    if (version) {
+        pstrcpy(s->version, sizeof(s->version), version);
+    } else {
+        pstrcpy(s->version, sizeof(s->version), QEMU_VERSION);
+    }
     ide_reset(s);
 }
 
@@ -2635,13 +2640,14 @@ void ide_init2(IDEBus *bus, DriveInfo *hd0, DriveInfo *hd1,
         s->unit = i;
         s->drive_serial = drive_serial++;
         s->io_buffer = qemu_blockalign(s->bs, IDE_DMA_BUF_SECTORS*512 + 4);
+        s->io_buffer_total_len = IDE_DMA_BUF_SECTORS*512 + 4;
         s->smart_selftest_data = qemu_blockalign(s->bs, 512);
         s->sector_write_timer = qemu_new_timer(vm_clock,
                                                ide_sector_write_timer_cb, s);
         if (i == 0)
-            ide_init_drive(s, hd0);
+            ide_init_drive(s, hd0, NULL);
         if (i == 1)
-            ide_init_drive(s, hd1);
+            ide_init_drive(s, hd1, NULL);
     }
     bus->irq = irq;
 }
@@ -2669,6 +2675,25 @@ static bool is_identify_set(void *opaque, int version_id)
     return s->identify_set != 0;
 }
 
+static EndTransferFunc* transfer_end_table[] = {
+        ide_sector_read,
+        ide_sector_write,
+        ide_transfer_stop,
+        ide_atapi_cmd_reply_end,
+        ide_atapi_cmd,
+};
+
+static int transfer_end_table_idx(EndTransferFunc *fn)
+{
+    int i;
+
+    for (i = 0; i < ARRAY_SIZE(transfer_end_table); i++)
+        if (transfer_end_table[i] == fn)
+            return i;
+
+    return -1;
+}
+
 static int ide_drive_post_load(void *opaque, int version_id)
 {
     IDEState *s = opaque;
@@ -2679,14 +2704,42 @@ static int ide_drive_post_load(void *opaque, int version_id)
             s->cdrom_changed = 1;
         }
     }
+
+    if (s->cur_io_buffer_len) {
+        s->end_transfer_func = transfer_end_table[s->end_transfer_fn_idx];
+        s->data_ptr = s->io_buffer + s->cur_io_buffer_offset;
+        s->data_end = s->data_ptr + s->cur_io_buffer_len;
+    }
+        
     return 0;
 }
 
+static void ide_drive_pre_save(void *opaque)
+{
+    IDEState *s = opaque;
+
+    s->cur_io_buffer_len = 0;
+
+    if (!(s->status & DRQ_STAT))
+        return;
+
+    s->cur_io_buffer_offset = s->data_ptr - s->io_buffer;
+    s->cur_io_buffer_len = s->data_end - s->data_ptr;
+
+    s->end_transfer_fn_idx = transfer_end_table_idx(s->end_transfer_func);
+    if (s->end_transfer_fn_idx == -1) {
+        fprintf(stderr, "%s: invalid end_transfer_func for DRQ_STAT\n",
+                        __func__);
+        s->end_transfer_fn_idx = 2;
+    }
+}
+
 const VMStateDescription vmstate_ide_drive = {
     .name = "ide_drive",
-    .version_id = 3,
+    .version_id = 4,
     .minimum_version_id = 0,
     .minimum_version_id_old = 0,
+    .pre_save = ide_drive_pre_save,
     .post_load = ide_drive_post_load,
     .fields      = (VMStateField []) {
         VMSTATE_INT32(mult_sectors, IDEState),
@@ -2709,7 +2762,14 @@ const VMStateDescription vmstate_ide_drive = {
         VMSTATE_UINT8(sense_key, IDEState),
         VMSTATE_UINT8(asc, IDEState),
         VMSTATE_UINT8_V(cdrom_changed, IDEState, 3),
-        /* XXX: if a transfer is pending, we do not save it yet */
+        VMSTATE_INT32_V(req_nb_sectors, IDEState, 4),
+        VMSTATE_VARRAY_INT32(io_buffer, IDEState, io_buffer_total_len, 4,
+			     vmstate_info_uint8, uint8_t),
+        VMSTATE_INT32_V(cur_io_buffer_offset, IDEState, 4),
+        VMSTATE_INT32_V(cur_io_buffer_len, IDEState, 4),
+        VMSTATE_UINT8_V(end_transfer_fn_idx, IDEState, 4),
+        VMSTATE_INT32_V(elementary_transfer_size, IDEState, 4),
+        VMSTATE_INT32_V(packet_transfer_size, IDEState, 4),
         VMSTATE_END_OF_LIST()
     }
 };
@@ -2767,10 +2827,6 @@ static void ide_dma_restart(IDEState *s, int is_read)
 void ide_dma_cancel(BMDMAState *bm)
 {
     if (bm->status & BM_STATUS_DMAING) {
-        bm->status &= ~BM_STATUS_DMAING;
-        /* cancel DMA request */
-        bm->unit = -1;
-        bm->dma_cb = NULL;
         if (bm->aiocb) {
 #ifdef DEBUG_AIO
             printf("aio_cancel\n");
@@ -2778,6 +2834,10 @@ void ide_dma_cancel(BMDMAState *bm)
             bdrv_aio_cancel(bm->aiocb);
             bm->aiocb = NULL;
         }
+        bm->status &= ~BM_STATUS_DMAING;
+        /* cancel DMA request */
+        bm->unit = -1;
+        bm->dma_cb = NULL;
     }
 }
 

hw/ide/internal.h

@@ -397,6 +397,7 @@ struct IDEState {
     /* set for lba48 access */
     uint8_t lba48;
     BlockDriverState *bs;
+    char version[9];
     /* ATAPI specific */
     uint8_t sense_key;
     uint8_t asc;
@@ -416,6 +417,11 @@ struct IDEState {
     uint8_t *data_ptr;
     uint8_t *data_end;
     uint8_t *io_buffer;
+    /* PIO save/restore */
+    int32_t io_buffer_total_len;
+    int cur_io_buffer_offset;
+    int cur_io_buffer_len;
+    uint8_t end_transfer_fn_idx;
     QEMUTimer *sector_write_timer; /* only used for win2k install hack */
     uint32_t irq_count; /* counts IRQs when using win2k install hack */
     /* CF-ATA extended error */
@@ -449,6 +455,7 @@ struct IDEDevice {
     DeviceState qdev;
     uint32_t unit;
     DriveInfo *dinfo;
+    char *version;
 };
 
 typedef int (*ide_qdev_initfn)(IDEDevice *dev);
@@ -474,6 +481,7 @@ struct BMDMAState {
     uint8_t status;
     uint32_t addr;
 
+    struct PCIIDEState *pci_dev;
     IDEBus *bus;
     /* current transfer state */
     uint32_t cur_addr;
@@ -548,7 +556,7 @@ uint32_t ide_data_readw(void *opaque, uint32_t addr);
 void ide_data_writel(void *opaque, uint32_t addr, uint32_t val);
 uint32_t ide_data_readl(void *opaque, uint32_t addr);
 
-void ide_init_drive(IDEState *s, DriveInfo *dinfo);
+void ide_init_drive(IDEState *s, DriveInfo *dinfo, const char *version);
 void ide_init2(IDEBus *bus, DriveInfo *hd0, DriveInfo *hd1,
                qemu_irq irq);
 void ide_init_ioport(IDEBus *bus, int iobase, int iobase2);

hw/ide/piix.c

@@ -78,6 +78,7 @@ static void bmdma_map(PCIDevice *pci_dev, int region_num,
         BMDMAState *bm = &d->bmdma[i];
         d->bus[i].bmdma = bm;
         bm->bus = d->bus+i;
+        bm->pci_dev = d;
         qemu_add_vm_change_state_handler(ide_dma_restart_cb, bm);
 
         register_ioport_write(addr, 1, 1, bmdma_cmd_writeb, bm);
@@ -161,7 +162,7 @@ void pci_piix3_ide_init(PCIBus *bus, DriveInfo **hd_table, int devfn)
 {
     PCIDevice *dev;
 
-    dev = pci_create_simple(bus, devfn, "PIIX3 IDE");
+    dev = pci_create_simple(bus, devfn, "piix3-ide");
     pci_ide_create_devs(dev, hd_table);
 }
 
@@ -171,18 +172,18 @@ void pci_piix4_ide_init(PCIBus *bus, DriveInfo **hd_table, int devfn)
 {
     PCIDevice *dev;
 
-    dev = pci_create_simple(bus, devfn, "PIIX4 IDE");
+    dev = pci_create_simple(bus, devfn, "piix4-ide");
     pci_ide_create_devs(dev, hd_table);
 }
 
 static PCIDeviceInfo piix_ide_info[] = {
     {
-        .qdev.name    = "PIIX3 IDE",
+        .qdev.name    = "piix3-ide",
         .qdev.size    = sizeof(PCIIDEState),
         .qdev.no_user = 1,
         .init         = pci_piix3_ide_initfn,
     },{
-        .qdev.name    = "PIIX4 IDE",
+        .qdev.name    = "piix4-ide",
         .qdev.size    = sizeof(PCIIDEState),
         .qdev.no_user = 1,
         .init         = pci_piix4_ide_initfn,

hw/ide/qdev.c

@@ -99,7 +99,7 @@ typedef struct IDEDrive {
 static int ide_drive_initfn(IDEDevice *dev)
 {
     IDEBus *bus = DO_UPCAST(IDEBus, qbus, dev->qdev.parent_bus);
-    ide_init_drive(bus->ifs + dev->unit, dev->dinfo);
+    ide_init_drive(bus->ifs + dev->unit, dev->dinfo, dev->version);
     return 0;
 }
 
@@ -110,6 +110,7 @@ static IDEDeviceInfo ide_drive_info = {
     .qdev.props = (Property[]) {
         DEFINE_PROP_UINT32("unit", IDEDrive, dev.unit, -1),
         DEFINE_PROP_DRIVE("drive", IDEDrive, dev.dinfo),
+        DEFINE_PROP_STRING("ver",  IDEDrive, dev.version),
         DEFINE_PROP_END_OF_LIST(),
     }
 };

hw/ioapic.c

@@ -22,12 +22,16 @@
 
 #include "hw.h"
 #include "pc.h"
+#include "sysemu.h"
 #include "qemu-timer.h"
 #include "host-utils.h"
 
+#include "qemu-kvm.h"
+
 //#define DEBUG_IOAPIC
 
 #define IOAPIC_NUM_PINS			0x18
+#define IOAPIC_DEFAULT_BASE_ADDRESS  0xfec00000
 #define IOAPIC_LVT_MASKED 		(1<<16)
 
 #define IOAPIC_TRIGGER_EDGE		0
@@ -45,6 +49,7 @@
 struct IOAPICState {
     uint8_t id;
     uint8_t ioregsel;
+    uint64_t base_address;
 
     uint32_t irr;
     uint64_t ioredtbl[IOAPIC_NUM_PINS];
@@ -94,8 +99,9 @@ void ioapic_set_irq(void *opaque, int vector, int level)
      * to GSI 2.  GSI maps to ioapic 1-1.  This is not
      * the cleanest way of doing it but it should work. */
 
-    if (vector == 0)
+    if (vector == 0 && irq0override) {
         vector = 2;
+    }
 
     if (vector >= 0 && vector < IOAPIC_NUM_PINS) {
         uint32_t mask = 1 << vector;
@@ -191,14 +197,91 @@ static void ioapic_mem_writel(void *opaque, target_phys_addr_t addr, uint32_t va
     }
 }
 
+static void kvm_kernel_ioapic_save_to_user(IOAPICState *s)
+{
+#if defined(KVM_CAP_IRQCHIP) && defined(TARGET_I386)
+    struct kvm_irqchip chip;
+    struct kvm_ioapic_state *kioapic;
+    int i;
+
+    chip.chip_id = KVM_IRQCHIP_IOAPIC;
+    kvm_get_irqchip(kvm_context, &chip);
+    kioapic = &chip.chip.ioapic;
+
+    s->id = kioapic->id;
+    s->ioregsel = kioapic->ioregsel;
+    s->base_address = kioapic->base_address;
+    s->irr = kioapic->irr;
+    for (i = 0; i < IOAPIC_NUM_PINS; i++) {
+        s->ioredtbl[i] = kioapic->redirtbl[i].bits;
+    }
+#endif
+}
+
+static void kvm_kernel_ioapic_load_from_user(IOAPICState *s)
+{
+#if defined(KVM_CAP_IRQCHIP) && defined(TARGET_I386)
+    struct kvm_irqchip chip;
+    struct kvm_ioapic_state *kioapic;
+    int i;
+
+    chip.chip_id = KVM_IRQCHIP_IOAPIC;
+    kioapic = &chip.chip.ioapic;
+
+    kioapic->id = s->id;
+    kioapic->ioregsel = s->ioregsel;
+    kioapic->base_address = s->base_address;
+    kioapic->irr = s->irr;
+    for (i = 0; i < IOAPIC_NUM_PINS; i++) {
+        kioapic->redirtbl[i].bits = s->ioredtbl[i];
+    }
+
+    kvm_set_irqchip(kvm_context, &chip);
+#endif
+}
+
+static void ioapic_pre_save(void *opaque)
+{
+    IOAPICState *s = (void *)opaque;
+ 
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_ioapic_save_to_user(s);
+    }
+}
+
+static int ioapic_pre_load(void *opaque)
+{
+    IOAPICState *s = opaque;
+
+    /* in case we are doing version 1, we just set these to sane values */
+    s->base_address = IOAPIC_DEFAULT_BASE_ADDRESS;
+    s->irr = 0;
+    return 0;
+}
+
+static int ioapic_post_load(void *opaque, int version_id)
+{
+    IOAPICState *s = opaque;
+
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_ioapic_load_from_user(s);
+    }
+    return 0;
+}
+
 static const VMStateDescription vmstate_ioapic = {
     .name = "ioapic",
-    .version_id = 1,
+    .version_id = 2,
     .minimum_version_id = 1,
     .minimum_version_id_old = 1,
+    .pre_load = ioapic_pre_load,
+    .post_load = ioapic_post_load,
+    .pre_save = ioapic_pre_save,
     .fields      = (VMStateField []) {
         VMSTATE_UINT8(id, IOAPICState),
         VMSTATE_UINT8(ioregsel, IOAPICState),
+        VMSTATE_UINT64_V(base_address, IOAPICState, 2),
+        VMSTATE_UINT32_V(irr, IOAPICState, 2),
         VMSTATE_UINT64_ARRAY(ioredtbl, IOAPICState, IOAPIC_NUM_PINS),
         VMSTATE_END_OF_LIST()
     }
@@ -210,8 +293,14 @@ static void ioapic_reset(void *opaque)
     int i;
 
     memset(s, 0, sizeof(*s));
+    s->base_address = IOAPIC_DEFAULT_BASE_ADDRESS;
     for(i = 0; i < IOAPIC_NUM_PINS; i++)
         s->ioredtbl[i] = 1 << 16; /* mask LVT */
+#ifdef KVM_CAP_IRQCHIP
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_kernel_ioapic_load_from_user(s);
+    }
+#endif
 }
 
 static CPUReadMemoryFunc * const ioapic_mem_read[3] = {

hw/ipf.c

@@ -0,0 +1,713 @@
+/*
+ * Itanium Platform Emulator derived from QEMU PC System Emulator
+ *
+ * Copyright (c) 2003-2004 Fabrice Bellard
+ *
+ * Copyright (c) 2007 Intel
+ * Ported for IA64 Platform Zhang Xiantao <xiantao.zhang@intel.com>
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining a copy
+ * of this software and associated documentation files (the "Software"), to deal
+ * in the Software without restriction, including without limitation the rights
+ * to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+ * copies of the Software, and to permit persons to whom the Software is
+ * furnished to do so, subject to the following conditions:
+ *
+ * The above copyright notice and this permission notice shall be included in
+ * all copies or substantial portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL
+ * THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+ * OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+ * THE SOFTWARE.
+ */
+
+#include "hw.h"
+#include "pc.h"
+#include "fdc.h"
+#include "pci.h"
+#include "block.h"
+#include "sysemu.h"
+#include "audio/audio.h"
+#include "net.h"
+#include "smbus.h"
+#include "boards.h"
+#include "firmware.h"
+#include "ia64intrin.h"
+#include <unistd.h>
+#include "device-assignment.h"
+#include "virtio-blk.h"
+
+#include "qemu-kvm.h"
+
+#define FW_FILENAME "Flash.fd"
+
+/* Leave a chunk of memory at the top of RAM for the BIOS ACPI tables.  */
+#define ACPI_DATA_SIZE       0x10000
+
+#define MAX_IDE_BUS 2
+
+static fdctrl_t *floppy_controller;
+static RTCState *rtc_state;
+static PCIDevice *i440fx_state;
+
+static uint32_t ipf_to_legacy_io(target_phys_addr_t addr)
+{
+    return (uint32_t)(((addr&0x3ffffff) >> 12 << 2)|((addr) & 0x3));
+}
+
+static void ipf_legacy_io_writeb(void *opaque, target_phys_addr_t addr,
+				 uint32_t val) {
+    uint32_t port = ipf_to_legacy_io(addr);
+
+    cpu_outb(0, port, val);
+}
+
+static void ipf_legacy_io_writew(void *opaque, target_phys_addr_t addr,
+				 uint32_t val) {
+    uint32_t port = ipf_to_legacy_io(addr);
+
+    cpu_outw(0, port, val);
+}
+
+static void ipf_legacy_io_writel(void *opaque, target_phys_addr_t addr,
+				 uint32_t val) {
+    uint32_t port = ipf_to_legacy_io(addr);
+
+    cpu_outl(0, port, val);
+}
+
+static uint32_t ipf_legacy_io_readb(void *opaque, target_phys_addr_t addr)
+{
+    uint32_t port = ipf_to_legacy_io(addr);
+
+    return cpu_inb(0, port);
+}
+
+static uint32_t ipf_legacy_io_readw(void *opaque, target_phys_addr_t addr)
+{
+    uint32_t port = ipf_to_legacy_io(addr);
+
+    return cpu_inw(0, port);
+}
+
+static uint32_t ipf_legacy_io_readl(void *opaque, target_phys_addr_t addr)
+{
+    uint32_t port = ipf_to_legacy_io(addr);
+
+    return cpu_inl(0, port);
+}
+
+static CPUReadMemoryFunc *ipf_legacy_io_read[3] = {
+    ipf_legacy_io_readb,
+    ipf_legacy_io_readw,
+    ipf_legacy_io_readl,
+};
+
+static CPUWriteMemoryFunc *ipf_legacy_io_write[3] = {
+    ipf_legacy_io_writeb,
+    ipf_legacy_io_writew,
+    ipf_legacy_io_writel,
+};
+
+static void pic_irq_request(void *opaque, int irq, int level)
+{
+    fprintf(stderr,"pic_irq_request called!\n");
+}
+
+/* PC cmos mappings */
+
+#define REG_EQUIPMENT_BYTE          0x14
+
+static int cmos_get_fd_drive_type(int fd0)
+{
+    int val;
+
+    switch (fd0) {
+    case 0:
+        /* 1.44 Mb 3"5 drive */
+        val = 4;
+        break;
+    case 1:
+        /* 2.88 Mb 3"5 drive */
+        val = 5;
+        break;
+    case 2:
+        /* 1.2 Mb 5"5 drive */
+        val = 2;
+        break;
+    default:
+        val = 0;
+        break;
+    }
+    return val;
+}
+
+static void cmos_init_hd(int type_ofs, int info_ofs, BlockDriverState *hd)
+{
+    RTCState *s = rtc_state;
+    int cylinders, heads, sectors;
+
+    bdrv_get_geometry_hint(hd, &cylinders, &heads, &sectors);
+    rtc_set_memory(s, type_ofs, 47);
+    rtc_set_memory(s, info_ofs, cylinders);
+    rtc_set_memory(s, info_ofs + 1, cylinders >> 8);
+    rtc_set_memory(s, info_ofs + 2, heads);
+    rtc_set_memory(s, info_ofs + 3, 0xff);
+    rtc_set_memory(s, info_ofs + 4, 0xff);
+    rtc_set_memory(s, info_ofs + 5, 0xc0 | ((heads > 8) << 3));
+    rtc_set_memory(s, info_ofs + 6, cylinders);
+    rtc_set_memory(s, info_ofs + 7, cylinders >> 8);
+    rtc_set_memory(s, info_ofs + 8, sectors);
+}
+
+/* convert boot_device letter to something recognizable by the bios */
+static int boot_device2nibble(char boot_device)
+{
+    switch(boot_device) {
+    case 'a':
+    case 'b':
+        return 0x01; /* floppy boot */
+    case 'c':
+        return 0x02; /* hard drive boot */
+    case 'd':
+        return 0x03; /* CD-ROM boot */
+    case 'n':
+        return 0x04; /* Network boot */
+    }
+    return 0;
+}
+
+/* hd_table must contain 4 block drivers */
+static void cmos_init(ram_addr_t ram_size, ram_addr_t above_4g_mem_size,
+                      const char *boot_device, BlockDriverState **hd_table)
+{
+    RTCState *s = rtc_state;
+    int nbds, bds[3] = { 0, };
+    int val;
+    int fd0, fd1, nb;
+    int i;
+
+    /* various important CMOS locations needed by PC/Bochs bios */
+
+    /* memory size */
+    val = 640; /* base memory in K */
+    rtc_set_memory(s, 0x15, val);
+    rtc_set_memory(s, 0x16, val >> 8);
+
+    val = (ram_size / 1024) - 1024;
+    if (val > 65535)
+        val = 65535;
+    rtc_set_memory(s, 0x17, val);
+    rtc_set_memory(s, 0x18, val >> 8);
+    rtc_set_memory(s, 0x30, val);
+    rtc_set_memory(s, 0x31, val >> 8);
+
+    if (above_4g_mem_size) {
+        rtc_set_memory(s, 0x5b, (unsigned int)above_4g_mem_size >> 16);
+        rtc_set_memory(s, 0x5c, (unsigned int)above_4g_mem_size >> 24);
+        rtc_set_memory(s, 0x5d, above_4g_mem_size >> 32);
+    }
+    rtc_set_memory(s, 0x5f, smp_cpus - 1);
+
+    if (ram_size > (16 * 1024 * 1024))
+        val = (ram_size / 65536) - ((16 * 1024 * 1024) / 65536);
+    else
+        val = 0;
+    if (val > 65535)
+        val = 65535;
+    rtc_set_memory(s, 0x34, val);
+    rtc_set_memory(s, 0x35, val >> 8);
+
+    /* set boot devices, and disable floppy signature check if requested */
+#define PC_MAX_BOOT_DEVICES 3
+    nbds = strlen(boot_device);
+
+    if (nbds > PC_MAX_BOOT_DEVICES) {
+        fprintf(stderr, "Too many boot devices for PC\n");
+        exit(1);
+    }
+
+    for (i = 0; i < nbds; i++) {
+        bds[i] = boot_device2nibble(boot_device[i]);
+        if (bds[i] == 0) {
+            fprintf(stderr, "Invalid boot device for PC: '%c'\n",
+                    boot_device[i]);
+            exit(1);
+        }
+    }
+
+    rtc_set_memory(s, 0x3d, (bds[1] << 4) | bds[0]);
+    rtc_set_memory(s, 0x38, (bds[2] << 4) | (fd_bootchk ?  0x0 : 0x1));
+
+    /* floppy type */
+
+    fd0 = fdctrl_get_drive_type(floppy_controller, 0);
+    fd1 = fdctrl_get_drive_type(floppy_controller, 1);
+
+    val = (cmos_get_fd_drive_type(fd0) << 4) | cmos_get_fd_drive_type(fd1);
+    rtc_set_memory(s, 0x10, val);
+
+    val = 0;
+    nb = 0;
+    if (fd0 < 3)
+        nb++;
+    if (fd1 < 3)
+        nb++;
+
+    switch (nb) {
+    case 0:
+        break;
+    case 1:
+        val |= 0x01; /* 1 drive, ready for boot */
+        break;
+    case 2:
+        val |= 0x41; /* 2 drives, ready for boot */
+        break;
+    }
+
+    val |= 0x02; /* FPU is there */
+    val |= 0x04; /* PS/2 mouse installed */
+    rtc_set_memory(s, REG_EQUIPMENT_BYTE, val);
+
+    /* hard drives */
+
+    rtc_set_memory(s, 0x12, (hd_table[0] ? 0xf0 : 0) | (hd_table[1] ? 0x0f : 0));
+    if (hd_table[0])
+        cmos_init_hd(0x19, 0x1b, hd_table[0]);
+    if (hd_table[1])
+        cmos_init_hd(0x1a, 0x24, hd_table[1]);
+
+    val = 0;
+    for (i = 0; i < 4; i++) {
+        if (hd_table[i]) {
+            int cylinders, heads, sectors, translation;
+            /* NOTE: bdrv_get_geometry_hint() returns the physical
+               geometry.  It is always such that: 1 <= sects <= 63, 1
+               <= heads <= 16, 1 <= cylinders <= 16383. The BIOS
+               geometry can be different if a translation is done. */
+            translation = bdrv_get_translation_hint(hd_table[i]);
+            if (translation == BIOS_ATA_TRANSLATION_AUTO) {
+                bdrv_get_geometry_hint(hd_table[i], &cylinders,
+                                       &heads, &sectors);
+                if (cylinders <= 1024 && heads <= 16 && sectors <= 63) {
+                    /* No translation. */
+                    translation = 0;
+                } else {
+                    /* LBA translation. */
+                    translation = 1;
+                }
+            } else {
+                translation--;
+            }
+            val |= translation << (i * 2);
+        }
+    }
+    rtc_set_memory(s, 0x39, val);
+}
+
+static void main_cpu_reset(void *opaque)
+{
+    CPUState *env = opaque;
+    cpu_reset(env);
+}
+
+static const int ide_iobase[2] = { 0x1f0, 0x170 };
+static const int ide_iobase2[2] = { 0x3f6, 0x376 };
+static const int ide_irq[2] = { 14, 15 };
+
+#define NE2000_NB_MAX 6
+
+static int ne2000_io[NE2000_NB_MAX] = { 0x300, 0x320, 0x340,
+                                        0x360, 0x280, 0x380 };
+static int ne2000_irq[NE2000_NB_MAX] = { 9, 10, 11, 3, 4, 5 };
+
+static int serial_io[MAX_SERIAL_PORTS] = { 0x3f8, 0x2f8, 0x3e8, 0x2e8 };
+static int serial_irq[MAX_SERIAL_PORTS] = { 4, 3, 4, 3 };
+
+static int parallel_io[MAX_PARALLEL_PORTS] = { 0x378, 0x278, 0x3bc };
+static int parallel_irq[MAX_PARALLEL_PORTS] = { 7, 7, 7 };
+
+#ifdef HAS_AUDIO
+static void audio_init (PCIBus *pci_bus, qemu_irq *pic)
+{
+    struct soundhw *c;
+    int audio_enabled = 0;
+
+    for (c = soundhw; !audio_enabled && c->name; ++c) {
+        audio_enabled = c->enabled;
+    }
+
+    if (audio_enabled) {
+        AudioState *s;
+
+        s = AUD_init ();
+        if (s) {
+            for (c = soundhw; c->name; ++c) {
+                if (c->enabled) {
+                    if (c->isa) {
+                        c->init.init_isa (s, pic);
+                    } else {
+                        if (pci_bus) {
+                            c->init.init_pci (pci_bus, s);
+                        }
+                    }
+                }
+            }
+        }
+    }
+}
+#endif
+
+static void pc_init_ne2k_isa(NICInfo *nd, qemu_irq *pic)
+{
+    static int nb_ne2k = 0;
+
+    if (nb_ne2k == NE2000_NB_MAX)
+        return;
+    isa_ne2000_init(ne2000_io[nb_ne2k], pic[ne2000_irq[nb_ne2k]], nd);
+    nb_ne2k++;
+}
+
+/* Itanium hardware initialisation */
+static void ipf_init1(ram_addr_t ram_size,
+                      const char *boot_device, DisplayState *ds,
+                      const char *kernel_filename, const char *kernel_cmdline,
+                      const char *initrd_filename,
+                      int pci_enabled, const char *cpu_model)
+{
+    char buf[1024];
+    int i;
+    ram_addr_t ram_addr;
+    ram_addr_t above_4g_mem_size = 0;
+    PCIBus *pci_bus;
+    PCIDevice *pci_dev;
+    int piix3_devfn = -1;
+    CPUState *env;
+    qemu_irq *cpu_irq;
+    qemu_irq *i8259;
+    int page_size;
+    int index;
+    unsigned long ipf_legacy_io_base, ipf_legacy_io_mem;
+    BlockDriverState *hd[MAX_IDE_BUS * MAX_IDE_DEVS];
+    BlockDriverState *fd[MAX_FD];
+
+    page_size = getpagesize();
+    if (page_size != TARGET_PAGE_SIZE) {
+	fprintf(stderr,"Error! Host page size != qemu target page size,"
+                " you may need to change TARGET_PAGE_BITS in qemu!"
+                "host page size:0x%x\n", page_size);
+        exit(-1);
+    };
+
+    if (ram_size >= 0xc0000000 ) {
+        above_4g_mem_size = ram_size - 0xc0000000;
+        ram_size = 0xc0000000;
+    }
+
+    /* init CPUs */
+    if (cpu_model == NULL) {
+        cpu_model = "IA64";
+    }
+
+    for(i = 0; i < smp_cpus; i++) {
+        env = cpu_init(cpu_model);
+        if (!env) {
+            fprintf(stderr, "Unable to find CPU definition\n");
+            exit(1);
+        }
+        if (i != 0)
+            env->hflags |= HF_HALTED_MASK;
+        register_savevm("cpu", i, 4, cpu_save, cpu_load, env);
+        qemu_register_reset(main_cpu_reset, 0, env);
+    }
+
+    /* allocate RAM */
+    if (kvm_enabled()) {
+        ram_addr = qemu_ram_alloc(0xa0000);
+        cpu_register_physical_memory(0, 0xa0000, ram_addr);
+
+        ram_addr = qemu_ram_alloc(0x20000); // Workaround 0xa0000-0xc0000
+
+        ram_addr = qemu_ram_alloc(0x40000);
+        cpu_register_physical_memory(0xc0000, 0x40000, ram_addr);
+
+        ram_addr = qemu_ram_alloc(ram_size - 0x100000);
+        cpu_register_physical_memory(0x100000, ram_size - 0x100000, ram_addr);
+    } else {
+        ram_addr = qemu_ram_alloc(ram_size);
+        cpu_register_physical_memory(0, ram_size, ram_addr);
+    }
+
+    /* above 4giga memory allocation */
+    if (above_4g_mem_size > 0) {
+        ram_addr = qemu_ram_alloc(above_4g_mem_size);
+        cpu_register_physical_memory(0x100000000, above_4g_mem_size, ram_addr);
+    }
+
+    /*Load firware to its proper position.*/
+    if (kvm_enabled()) {
+        unsigned long  image_size;
+        uint8_t *image = NULL;
+        unsigned long nvram_addr;
+        unsigned long nvram_fd = 0;
+        unsigned long type = READ_FROM_NVRAM;
+        unsigned long i = 0;
+        unsigned long fw_offset;
+        ram_addr_t fw_mem = qemu_ram_alloc(GFW_SIZE);
+
+        snprintf(buf, sizeof(buf), "%s/%s", bios_dir, FW_FILENAME);
+        image = read_image(buf, &image_size );
+        if (NULL == image || !image_size) {
+            fprintf(stderr, "Error when reading Guest Firmware!\n");
+            fprintf(stderr, "Please check Guest firmware at %s\n", buf);
+            exit(1);
+        }
+        fw_offset = GFW_START + GFW_SIZE - image_size;
+
+        cpu_register_physical_memory(GFW_START, GFW_SIZE, fw_mem);
+        cpu_physical_memory_write(fw_offset, image, image_size);
+
+        free(image);
+
+        if (nvram) {
+            nvram_addr = NVRAM_START;
+            nvram_fd = kvm_ia64_nvram_init(type);
+            if (nvram_fd != -1) {
+                kvm_ia64_copy_from_nvram_to_GFW(nvram_fd);
+                close(nvram_fd);
+            }
+            i = atexit((void *)kvm_ia64_copy_from_GFW_to_nvram);
+            if (i != 0)
+                fprintf(stderr, "cannot set exit function\n");
+        } else
+            nvram_addr = 0;
+
+        kvm_ia64_build_hob(ram_size + above_4g_mem_size, smp_cpus, nvram_addr);
+    }
+
+    /*Register legacy io address space, size:64M*/
+    ipf_legacy_io_base = 0xE0000000;
+    ipf_legacy_io_mem = cpu_register_io_memory(0, ipf_legacy_io_read,
+                                               ipf_legacy_io_write, NULL);
+    cpu_register_physical_memory(ipf_legacy_io_base, 64*1024*1024,
+                                 ipf_legacy_io_mem);
+
+    cpu_irq = qemu_allocate_irqs(pic_irq_request, first_cpu, 1);
+    i8259 = kvm_i8259_init(cpu_irq[0]);
+
+    if (pci_enabled) {
+        pci_bus = i440fx_init(&i440fx_state, i8259);
+        piix3_devfn = piix3_init(pci_bus, -1);
+    } else {
+        pci_bus = NULL;
+    }
+
+    if (cirrus_vga_enabled) {
+        if (pci_enabled)
+            pci_cirrus_vga_init(pci_bus);
+        else
+            isa_cirrus_vga_init();
+    } else {
+        if (pci_enabled)
+            pci_vga_init(pci_bus, 0, 0);
+        else
+            isa_vga_init();
+    }
+
+    rtc_state = rtc_init(0x70, i8259[8], 2000);
+
+    if (pci_enabled) {
+        pic_set_alt_irq_func(isa_pic, NULL, NULL);
+    }
+
+    for(i = 0; i < MAX_SERIAL_PORTS; i++) {
+        if (serial_hds[i]) {
+            serial_init(serial_io[i], i8259[serial_irq[i]], 115200,
+                        serial_hds[i]);
+        }
+    }
+
+    for(i = 0; i < MAX_PARALLEL_PORTS; i++) {
+        if (parallel_hds[i]) {
+            parallel_init(parallel_io[i], i8259[parallel_irq[i]],
+                          parallel_hds[i]);
+        }
+    }
+
+    for(i = 0; i < nb_nics; i++) {
+        NICInfo *nd = &nd_table[i];
+
+        if (!pci_enabled || (nd->model && strcmp(nd->model, "ne2k_isa") == 0))
+            pc_init_ne2k_isa(nd, i8259);
+        else
+            pci_nic_init(nd, "e1000", NULL);
+    }
+
+#undef USE_HYPERCALL  //Disable it now, need to implement later!
+#ifdef USE_HYPERCALL
+    pci_hypercall_init(pci_bus);
+#endif
+
+    if (drive_get_max_bus(IF_IDE) >= MAX_IDE_BUS) {
+        fprintf(stderr, "qemu: too many IDE bus\n");
+        exit(1);
+    }
+
+    for(i = 0; i < MAX_IDE_BUS * MAX_IDE_DEVS; i++) {
+        index = drive_get_index(IF_IDE, i / MAX_IDE_DEVS, i % MAX_IDE_DEVS);
+	if (index != -1)
+	    hd[i] = drives_table[index].bdrv;
+	else
+	    hd[i] = NULL;
+    }
+
+    if (pci_enabled) {
+        pci_piix3_ide_init(pci_bus, hd, piix3_devfn + 1, i8259);
+    } else {
+        for(i = 0; i < MAX_IDE_BUS; i++) {
+            isa_ide_init(ide_iobase[i], ide_iobase2[i], i8259[ide_irq[i]],
+	                 hd[MAX_IDE_DEVS * i], hd[MAX_IDE_DEVS * i + 1]);
+        }
+    }
+
+    i8042_init(i8259[1], i8259[12], 0x60);
+    DMA_init(0);
+#ifdef HAS_AUDIO
+    audio_init(pci_enabled ? pci_bus : NULL, i8259);
+#endif
+
+    for(i = 0; i < MAX_FD; i++) {
+        index = drive_get_index(IF_FLOPPY, 0, i);
+	if (index != -1)
+	    fd[i] = drives_table[index].bdrv;
+	else
+	    fd[i] = NULL;
+    }
+    floppy_controller = fdctrl_init(i8259[6], 2, 0, 0x3f0, fd);
+
+    cmos_init(ram_size, above_4g_mem_size, boot_device, hd);
+
+    if (pci_enabled && usb_enabled) {
+        usb_uhci_piix3_init(pci_bus, piix3_devfn + 2);
+    }
+
+    if (pci_enabled && acpi_enabled) {
+        uint8_t *eeprom_buf = qemu_mallocz(8 * 256); /* XXX: make this persistent */
+        i2c_bus *smbus;
+
+        /* TODO: Populate SPD eeprom data.  */
+        smbus = piix4_pm_init(pci_bus, piix3_devfn + 3, 0xb100, i8259[9]);
+        for (i = 0; i < 8; i++) {
+            DeviceState *eeprom;
+            eeprom = qdev_create((BusState *)smbus, "smbus-eeprom");
+            qdev_set_prop_int(eeprom, "address", 0x50 + i);
+            qdev_set_prop_ptr(eeprom, "data", eeprom_buf + (i * 256));
+            qdev_init(eeprom);
+        }
+    }
+
+    if (i440fx_state) {
+        i440fx_init_memory_mappings(i440fx_state);
+    }
+
+    if (pci_enabled) {
+	int max_bus;
+        int bus;
+
+        max_bus = drive_get_max_bus(IF_SCSI);
+	for (bus = 0; bus <= max_bus; bus++) {
+            pci_create_simple(pci_bus, -1, "lsi53c895a");
+        }
+    }
+    /* Add virtio block devices */
+    if (pci_enabled) {
+	int index;
+	int unit_id = 0;
+
+	while ((index = drive_get_index(IF_VIRTIO, 0, unit_id)) != -1) {
+            pci_dev = pci_create("virtio-blk-pci",
+                                 drives_table[index].devaddr);
+            qdev_init(&pci_dev->qdev);
+	    unit_id++;
+	}
+    }
+
+#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
+    if (kvm_enabled())
+	add_assigned_devices(pci_bus, assigned_devices, assigned_devices_index);
+#endif /* CONFIG_KVM_DEVICE_ASSIGNMENT */
+
+}
+
+static void ipf_init_pci(ram_addr_t ram_size,
+                         const char *boot_device, DisplayState *ds,
+                         const char *kernel_filename,
+                         const char *kernel_cmdline,
+                         const char *initrd_filename,
+                         const char *cpu_model)
+{
+    ipf_init1(ram_size, boot_device, ds, kernel_filename,
+              kernel_cmdline, initrd_filename, 1, cpu_model);
+}
+
+QEMUMachine ipf_machine = {
+    .name = "itanium",
+    .desc = "Itanium Platform",
+    .init = (QEMUMachineInitFunc *)ipf_init_pci,
+    .max_cpus = 255,
+    .is_default = 1,
+};
+
+static void ipf_machine_init(void)
+{
+    qemu_register_machine(&ipf_machine);
+}
+
+machine_init(ipf_machine_init);
+
+#define IOAPIC_NUM_PINS 48
+
+static int ioapic_irq_count[IOAPIC_NUM_PINS];
+
+static int ioapic_map_irq(int devfn, int irq_num)
+{
+    int irq, dev;
+    dev = devfn >> 3;
+    irq = ((((dev << 2) + (dev >> 3) + irq_num) & 31) + 16);
+    return irq;
+}
+
+/*
+ * Dummy function to provide match for call from hw/apic.c
+ */
+void apic_set_irq_delivered(void) {
+}
+
+void ioapic_set_irq(void *opaque, int irq_num, int level)
+{
+    int vector, pic_ret;
+
+    PCIDevice *pci_dev = (PCIDevice *)opaque;
+    vector = ioapic_map_irq(pci_dev->devfn, irq_num);
+
+    if (level)
+        ioapic_irq_count[vector] += 1;
+    else
+        ioapic_irq_count[vector] -= 1;
+
+    if (kvm_enabled()) {
+	if (kvm_set_irq(vector, ioapic_irq_count[vector] == 0, &pic_ret))
+            if (pic_ret != 0)
+                apic_set_irq_delivered();
+	    return;
+    }
+}
+
+int ipf_map_irq(PCIDevice *pci_dev, int irq_num)
+{
+	return ioapic_map_irq(pci_dev->devfn, irq_num);
+}

hw/loader.c

@@ -48,6 +48,7 @@
 #include "sysemu.h"
 #include "uboot_image.h"
 #include "loader.h"
+#include "fw_cfg.h"
 
 #include <zlib.h>
 
@@ -526,15 +527,15 @@ struct Rom {
     char *path;
     size_t romsize;
     uint8_t *data;
-    int align;
     int isrom;
+    char *fw_dir;
+    char *fw_file;
 
-    target_phys_addr_t min;
-    target_phys_addr_t max;
     target_phys_addr_t addr;
     QTAILQ_ENTRY(Rom) next;
 };
 
+static FWCfgState *fw_cfg;
 static QTAILQ_HEAD(, Rom) roms = QTAILQ_HEAD_INITIALIZER(roms);
 int rom_enable_driver_roms;
 
@@ -548,7 +549,7 @@ static void rom_insert(Rom *rom)
 
     /* list is ordered by load address */
     QTAILQ_FOREACH(item, &roms, next) {
-        if (rom->min >= item->min)
+        if (rom->addr >= item->addr)
             continue;
         QTAILQ_INSERT_BEFORE(item, rom, next);
         return;
@@ -556,8 +557,8 @@ static void rom_insert(Rom *rom)
     QTAILQ_INSERT_TAIL(&roms, rom, next);
 }
 
-int rom_add_file(const char *file,
+int rom_add_file(const char *file, const char *fw_dir,
-                 target_phys_addr_t min, target_phys_addr_t max, int align)
+                 target_phys_addr_t addr)
 {
     Rom *rom;
     int rc, fd = -1;
@@ -576,9 +577,11 @@ int rom_add_file(const char *file,
         goto err;
     }
 
-    rom->align   = align;
+    if (fw_dir) {
-    rom->min     = min;
+        rom->fw_dir  = qemu_strdup(fw_dir);
-    rom->max     = max;
+        rom->fw_file = qemu_strdup(file);
+    }
+    rom->addr    = addr;
     rom->romsize = lseek(fd, 0, SEEK_END);
     rom->data    = qemu_mallocz(rom->romsize);
     lseek(fd, 0, SEEK_SET);
@@ -590,6 +593,8 @@ int rom_add_file(const char *file,
     }
     close(fd);
     rom_insert(rom);
+    if (rom->fw_file && fw_cfg)
+        fw_cfg_add_file(fw_cfg, rom->fw_dir, rom->fw_file, rom->data, rom->romsize);
     return 0;
 
 err:
@@ -603,15 +608,13 @@ err:
 }
 
 int rom_add_blob(const char *name, const void *blob, size_t len,
-                 target_phys_addr_t min, target_phys_addr_t max, int align)
+                 target_phys_addr_t addr)
 {
     Rom *rom;
 
     rom = qemu_mallocz(sizeof(*rom));
     rom->name    = qemu_strdup(name);
-    rom->align   = align;
+    rom->addr    = addr;
-    rom->min     = min;
-    rom->max     = max;
     rom->romsize = len;
     rom->data    = qemu_mallocz(rom->romsize);
     memcpy(rom->data, blob, len);
@@ -623,14 +626,14 @@ int rom_add_vga(const char *file)
 {
     if (!rom_enable_driver_roms)
         return 0;
-    return rom_add_file(file, PC_ROM_MIN_VGA, PC_ROM_MAX, PC_ROM_ALIGN);
+    return rom_add_file(file, "vgaroms", 0);
 }
 
 int rom_add_option(const char *file)
 {
     if (!rom_enable_driver_roms)
         return 0;
-    return rom_add_file(file, PC_ROM_MIN_OPTION, PC_ROM_MAX, PC_ROM_ALIGN);
+    return rom_add_file(file, "genroms", 0);
 }
 
 static void rom_reset(void *unused)
@@ -638,8 +641,12 @@ static void rom_reset(void *unused)
     Rom *rom;
 
     QTAILQ_FOREACH(rom, &roms, next) {
-        if (rom->data == NULL)
+        if (rom->fw_file) {
             continue;
+        }
+        if (rom->data == NULL) {
+            continue;
+        }
         cpu_physical_memory_write_rom(rom->addr, rom->data, rom->romsize);
         if (rom->isrom) {
             /* rom needs to be written only once */
@@ -656,32 +663,17 @@ int rom_load_all(void)
     Rom *rom;
 
     QTAILQ_FOREACH(rom, &roms, next) {
-        if (addr < rom->min)
+        if (rom->fw_file) {
-            addr = rom->min;
+            continue;
-        if (rom->max) {
-            /* load address range */
-            if (rom->align) {
-                addr += (rom->align-1);
-                addr &= ~(rom->align-1);
-            }
-            if (addr + rom->romsize > rom->max) {
-                fprintf(stderr, "rom: out of memory (rom %s, "
-                        "addr 0x" TARGET_FMT_plx
-                        ", size 0x%zx, max 0x" TARGET_FMT_plx ")\n",
-                        rom->name, addr, rom->romsize, rom->max);
-                return -1;
-            }
-        } else {
-            /* fixed address requested */
-            if (addr != rom->min) {
-                fprintf(stderr, "rom: requested regions overlap "
-                        "(rom %s. free=0x" TARGET_FMT_plx
-                        ", addr=0x" TARGET_FMT_plx ")\n",
-                        rom->name, addr, rom->min);
-                return -1;
-            }
         }
-        rom->addr = addr;
+        if (addr > rom->addr) {
+            fprintf(stderr, "rom: requested regions overlap "
+                    "(rom %s. free=0x" TARGET_FMT_plx
+                    ", addr=0x" TARGET_FMT_plx ")\n",
+                    rom->name, addr, rom->addr);
+            return -1;
+        }
+        addr  = rom->addr;
         addr += rom->romsize;
         memtype = cpu_get_physical_page_desc(rom->addr) & (3 << IO_MEM_SHIFT);
         if (memtype == IO_MEM_ROM)
@@ -692,22 +684,35 @@ int rom_load_all(void)
     return 0;
 }
 
+void rom_set_fw(void *f)
+{
+    fw_cfg = f;
+}
+
 static Rom *find_rom(target_phys_addr_t addr)
 {
     Rom *rom;
 
     QTAILQ_FOREACH(rom, &roms, next) {
-        if (rom->max)
+        if (rom->fw_file) {
             continue;
-        if (rom->min > addr)
+        }
+        if (rom->addr > addr) {
             continue;
-        if (rom->min + rom->romsize < addr)
+        }
+        if (rom->addr + rom->romsize < addr) {
             continue;
+        }
         return rom;
     }
     return NULL;
 }
 
+/*
+ * Copies memory from registered ROMs to dest. Any memory that is contained in
+ * a ROM between addr and addr + size is copied. Note that this can involve
+ * multiple ROMs, which need not start at addr and need not end at addr + size.
+ */
 int rom_copy(uint8_t *dest, target_phys_addr_t addr, size_t size)
 {
     target_phys_addr_t end = addr + size;
@@ -716,25 +721,27 @@ int rom_copy(uint8_t *dest, target_phys_addr_t addr, size_t size)
     Rom *rom;
 
     QTAILQ_FOREACH(rom, &roms, next) {
-        if (rom->max)
+        if (rom->fw_file) {
             continue;
-        if (rom->min > addr)
+        }
+        if (rom->addr + rom->romsize < addr) {
             continue;
-        if (rom->min + rom->romsize < addr)
+        }
-            continue;
+        if (rom->addr > end) {
-        if (rom->min > end)
             break;
-        if (!rom->data)
+        }
+        if (!rom->data) {
             continue;
+        }
 
-        d = dest + (rom->min - addr);
+        d = dest + (rom->addr - addr);
         s = rom->data;
         l = rom->romsize;
 
-        if (rom->min < addr) {
+        if (rom->addr < addr) {
             d = dest;
-            s += (addr - rom->min);
+            s += (addr - rom->addr);
-            l -= (addr - rom->min);
+            l -= (addr - rom->addr);
         }
         if ((d + l) > (dest + size)) {
             l = dest - d;
@@ -753,7 +760,7 @@ void *rom_ptr(target_phys_addr_t addr)
     rom = find_rom(addr);
     if (!rom || !rom->data)
         return NULL;
-    return rom->data + (addr - rom->min);
+    return rom->data + (addr - rom->addr);
 }
 
 void do_info_roms(Monitor *mon)
@@ -761,10 +768,19 @@ void do_info_roms(Monitor *mon)
     Rom *rom;
 
     QTAILQ_FOREACH(rom, &roms, next) {
-        monitor_printf(mon, "addr=" TARGET_FMT_plx
+        if (!rom->fw_file) {
-                       " size=0x%06zx mem=%s name=\"%s\" \n",
+            monitor_printf(mon, "addr=" TARGET_FMT_plx
-                       rom->addr, rom->romsize,
+                           " size=0x%06zx mem=%s name=\"%s\" \n",
-                       rom->isrom ? "rom" : "ram",
+                           rom->addr, rom->romsize,
-                       rom->name);
+                           rom->isrom ? "rom" : "ram",
+                           rom->name);
+        } else {
+            monitor_printf(mon, "fw=%s/%s"
+                           " size=0x%06zx name=\"%s\" \n",
+                           rom->fw_dir,
+                           rom->fw_file,
+                           rom->romsize,
+                           rom->name);
+        }
     }
 }

hw/loader.h

@@ -19,19 +19,21 @@ void pstrcpy_targphys(const char *name,
                       target_phys_addr_t dest, int buf_size,
                       const char *source);
 
-int rom_add_file(const char *file,
+
-                 target_phys_addr_t min, target_phys_addr_t max, int align);
+int rom_add_file(const char *file, const char *fw_dir,
+                 target_phys_addr_t addr);
 int rom_add_blob(const char *name, const void *blob, size_t len,
-                 target_phys_addr_t min, target_phys_addr_t max, int align);
+                 target_phys_addr_t addr);
 int rom_load_all(void);
+void rom_set_fw(void *f);
 int rom_copy(uint8_t *dest, target_phys_addr_t addr, size_t size);
 void *rom_ptr(target_phys_addr_t addr);
 void do_info_roms(Monitor *mon);
 
 #define rom_add_file_fixed(_f, _a)              \
-    rom_add_file(_f, _a, 0, 0)
+    rom_add_file(_f, NULL, _a)
 #define rom_add_blob_fixed(_f, _b, _l, _a)      \
-    rom_add_blob(_f, _b, _l, _a, 0, 0)
+    rom_add_blob(_f, _b, _l, _a)
 
 #define PC_ROM_MIN_VGA     0xc0000
 #define PC_ROM_MIN_OPTION  0xc8000

hw/lsi53c895a.c

@@ -173,11 +173,15 @@ do { fprintf(stderr, "lsi_scsi: error: " fmt , ## __VA_ARGS__);} while (0)
 /* Flag set if this is a tagged command.  */
 #define LSI_TAG_VALID     (1 << 16)
 
-typedef struct {
+typedef struct lsi_request {
     uint32_t tag;
+    SCSIDevice *dev;
+    uint32_t dma_len;
+    uint8_t *dma_buf;
     uint32_t pending;
     int out;
-} lsi_queue;
+    QTAILQ_ENTRY(lsi_request) next;
+} lsi_request;
 
 typedef struct {
     PCIDevice dev;
@@ -198,16 +202,13 @@ typedef struct {
      * 3 if a DMA operation is in progress.  */
     int waiting;
     SCSIBus bus;
-    SCSIDevice *current_dev;
+    SCSIDevice *select_dev;
     int current_lun;
     /* The tag is a combination of the device ID and the SCSI tag.  */
-    uint32_t current_tag;
+    uint32_t select_tag;
-    uint32_t current_dma_len;
     int command_complete;
-    uint8_t *dma_buf;
+    QTAILQ_HEAD(, lsi_request) queue;
-    lsi_queue *queue;
+    lsi_request *current;
-    int queue_len;
-    int active_commands;
 
     uint32_t dsa;
     uint32_t temp;
@@ -370,7 +371,7 @@ static int lsi_dma_64bit(LSIState *s)
 static uint8_t lsi_reg_readb(LSIState *s, int offset);
 static void lsi_reg_writeb(LSIState *s, int offset, uint8_t val);
 static void lsi_execute_script(LSIState *s);
-static void lsi_reselect(LSIState *s, uint32_t tag);
+static void lsi_reselect(LSIState *s, lsi_request *p);
 
 static inline uint32_t read_dword(LSIState *s, uint32_t addr)
 {
@@ -391,9 +392,9 @@ static void lsi_stop_script(LSIState *s)
 
 static void lsi_update_irq(LSIState *s)
 {
-    int i;
     int level;
     static int last_level;
+    lsi_request *p;
 
     /* It's unclear whether the DIP/SIP bits should be cleared when the
        Interrupt Status Registers are cleared or when istat0 is read.
@@ -427,9 +428,9 @@ static void lsi_update_irq(LSIState *s)
     if (!level && lsi_irq_on_rsl(s) && !(s->scntl1 & LSI_SCNTL1_CON)) {
         DPRINTF("Handled IRQs & disconnected, looking for pending "
                 "processes\n");
-        for (i = 0; i < s->active_commands; i++) {
+        QTAILQ_FOREACH(p, &s->queue, next) {
-            if (s->queue[i].pending) {
+            if (p->pending) {
-                lsi_reselect(s, s->queue[i].tag);
+                lsi_reselect(s, p);
                 break;
             }
         }
@@ -508,15 +509,16 @@ static void lsi_do_dma(LSIState *s, int out)
     uint32_t count;
     target_phys_addr_t addr;
 
-    if (!s->current_dma_len) {
+    assert(s->current);
+    if (!s->current->dma_len) {
         /* Wait until data is available.  */
         DPRINTF("DMA no data available\n");
         return;
     }
 
     count = s->dbc;
-    if (count > s->current_dma_len)
+    if (count > s->current->dma_len)
-        count = s->current_dma_len;
+        count = s->current->dma_len;
 
     addr = s->dnad;
     /* both 40 and Table Indirect 64-bit DMAs store upper bits in dnad64 */
@@ -532,29 +534,29 @@ static void lsi_do_dma(LSIState *s, int out)
     s->dnad += count;
     s->dbc -= count;
 
-    if (s->dma_buf == NULL) {
+    if (s->current->dma_buf == NULL) {
-        s->dma_buf = s->current_dev->info->get_buf(s->current_dev,
+        s->current->dma_buf = s->current->dev->info->get_buf(s->current->dev,
-                                                   s->current_tag);
+                                                             s->current->tag);
     }
 
     /* ??? Set SFBR to first data byte.  */
     if (out) {
-        cpu_physical_memory_read(addr, s->dma_buf, count);
+        cpu_physical_memory_read(addr, s->current->dma_buf, count);
     } else {
-        cpu_physical_memory_write(addr, s->dma_buf, count);
+        cpu_physical_memory_write(addr, s->current->dma_buf, count);
     }
-    s->current_dma_len -= count;
+    s->current->dma_len -= count;
-    if (s->current_dma_len == 0) {
+    if (s->current->dma_len == 0) {
-        s->dma_buf = NULL;
+        s->current->dma_buf = NULL;
         if (out) {
             /* Write the data.  */
-            s->current_dev->info->write_data(s->current_dev, s->current_tag);
+            s->current->dev->info->write_data(s->current->dev, s->current->tag);
         } else {
             /* Request any remaining data.  */
-            s->current_dev->info->read_data(s->current_dev, s->current_tag);
+            s->current->dev->info->read_data(s->current->dev, s->current->tag);
         }
     } else {
-        s->dma_buf += count;
+        s->current->dma_buf += count;
         lsi_resume_script(s);
     }
 }
@@ -563,15 +565,14 @@ static void lsi_do_dma(LSIState *s, int out)
 /* Add a command to the queue.  */
 static void lsi_queue_command(LSIState *s)
 {
-    lsi_queue *p;
+    lsi_request *p = s->current;
 
     DPRINTF("Queueing tag=0x%x\n", s->current_tag);
-    if (s->queue_len == s->active_commands) {
+    assert(s->current != NULL);
-        s->queue_len++;
+    assert(s->current->dma_len == 0);
-        s->queue = qemu_realloc(s->queue, s->queue_len * sizeof(lsi_queue));
+    QTAILQ_INSERT_TAIL(&s->queue, s->current, next);
-    }
+    s->current = NULL;
-    p = &s->queue[s->active_commands++];
+
-    p->tag = s->current_tag;
     p->pending = 0;
     p->out = (s->sstat1 & PHASE_MASK) == PHASE_DO;
 }
@@ -588,45 +589,29 @@ static void lsi_add_msg_byte(LSIState *s, uint8_t data)
 }
 
 /* Perform reselection to continue a command.  */
-static void lsi_reselect(LSIState *s, uint32_t tag)
+static void lsi_reselect(LSIState *s, lsi_request *p)
 {
-    lsi_queue *p;
-    int n;
     int id;
 
-    p = NULL;
+    assert(s->current == NULL);
-    for (n = 0; n < s->active_commands; n++) {
+    QTAILQ_REMOVE(&s->queue, p, next);
-        p = &s->queue[n];
+    s->current = p;
-        if (p->tag == tag)
+
-            break;
+    id = (p->tag >> 8) & 0xf;
-    }
-    if (n == s->active_commands) {
-        BADF("Reselected non-existant command tag=0x%x\n", tag);
-        return;
-    }
-    id = (tag >> 8) & 0xf;
     s->ssid = id | 0x80;
     /* LSI53C700 Family Compatibility, see LSI53C895A 4-73 */
     if (!s->dcntl & LSI_DCNTL_COM) {
         s->sfbr = 1 << (id & 0x7);
     }
     DPRINTF("Reselected target %d\n", id);
-    s->current_dev = s->bus.devs[id];
-    s->current_tag = tag;
     s->scntl1 |= LSI_SCNTL1_CON;
     lsi_set_phase(s, PHASE_MI);
     s->msg_action = p->out ? 2 : 3;
-    s->current_dma_len = p->pending;
+    s->current->dma_len = p->pending;
-    s->dma_buf = NULL;
     lsi_add_msg_byte(s, 0x80);
-    if (s->current_tag & LSI_TAG_VALID) {
+    if (s->current->tag & LSI_TAG_VALID) {
         lsi_add_msg_byte(s, 0x20);
-        lsi_add_msg_byte(s, tag & 0xff);
+        lsi_add_msg_byte(s, p->tag & 0xff);
-    }
-
-    s->active_commands--;
-    if (n != s->active_commands) {
-        s->queue[n] = s->queue[s->active_commands];
     }
 
     if (lsi_irq_on_rsl(s)) {
@@ -638,10 +623,9 @@ static void lsi_reselect(LSIState *s, uint32_t tag)
    the device was reselected, nonzero if the IO is deferred.  */
 static int lsi_queue_tag(LSIState *s, uint32_t tag, uint32_t arg)
 {
-    lsi_queue *p;
+    lsi_request *p;
-    int i;
+
-    for (i = 0; i < s->active_commands; i++) {
+    QTAILQ_FOREACH(p, &s->queue, next) {
-        p = &s->queue[i];
         if (p->tag == tag) {
             if (p->pending) {
                 BADF("Multiple IO pending for tag %d\n", tag);
@@ -656,10 +640,10 @@ static int lsi_queue_tag(LSIState *s, uint32_t tag, uint32_t arg)
                 (lsi_irq_on_rsl(s) && !(s->scntl1 & LSI_SCNTL1_CON) &&
                  !(s->istat0 & (LSI_ISTAT0_SIP | LSI_ISTAT0_DIP)))) {
                 /* Reselect device.  */
-                lsi_reselect(s, tag);
+                lsi_reselect(s, p);
                 return 0;
             } else {
-               DPRINTF("Queueing IO tag=0x%x\n", tag);
+                DPRINTF("Queueing IO tag=0x%x\n", tag);
                 p->pending = arg;
                 return 1;
             }
@@ -687,11 +671,15 @@ static void lsi_command_complete(SCSIBus *bus, int reason, uint32_t tag,
         } else {
             lsi_set_phase(s, PHASE_ST);
         }
+
+        qemu_free(s->current);
+        s->current = NULL;
+
         lsi_resume_script(s);
         return;
     }
 
-    if (s->waiting == 1 || tag != s->current_tag ||
+    if (s->waiting == 1 || !s->current || tag != s->current->tag ||
         (lsi_irq_on_rsl(s) && !(s->scntl1 & LSI_SCNTL1_CON))) {
         if (lsi_queue_tag(s, tag, arg))
             return;
@@ -699,7 +687,7 @@ static void lsi_command_complete(SCSIBus *bus, int reason, uint32_t tag,
 
     /* host adapter (re)connected */
     DPRINTF("Data ready tag=0x%x len=%d\n", tag, arg);
-    s->current_dma_len = arg;
+    s->current->dma_len = arg;
     s->command_complete = 1;
     if (!s->waiting)
         return;
@@ -721,14 +709,20 @@ static void lsi_do_command(LSIState *s)
     cpu_physical_memory_read(s->dnad, buf, s->dbc);
     s->sfbr = buf[0];
     s->command_complete = 0;
-    n = s->current_dev->info->send_command(s->current_dev, s->current_tag, buf,
+
-                                           s->current_lun);
+    assert(s->current == NULL);
+    s->current = qemu_mallocz(sizeof(lsi_request));
+    s->current->tag = s->select_tag;
+    s->current->dev = s->select_dev;
+
+    n = s->current->dev->info->send_command(s->current->dev, s->current->tag, buf,
+                                            s->current_lun);
     if (n > 0) {
         lsi_set_phase(s, PHASE_DI);
-        s->current_dev->info->read_data(s->current_dev, s->current_tag);
+        s->current->dev->info->read_data(s->current->dev, s->current->tag);
     } else if (n < 0) {
         lsi_set_phase(s, PHASE_DO);
-        s->current_dev->info->write_data(s->current_dev, s->current_tag);
+        s->current->dev->info->write_data(s->current->dev, s->current->tag);
     }
 
     if (!s->command_complete) {
@@ -851,16 +845,16 @@ static void lsi_do_msgout(LSIState *s)
             }
             break;
         case 0x20: /* SIMPLE queue */
-            s->current_tag |= lsi_get_msgbyte(s) | LSI_TAG_VALID;
+            s->select_tag |= lsi_get_msgbyte(s) | LSI_TAG_VALID;
             DPRINTF("SIMPLE queue tag=0x%x\n", s->current_tag & 0xff);
             break;
         case 0x21: /* HEAD of queue */
             BADF("HEAD queue not implemented\n");
-            s->current_tag |= lsi_get_msgbyte(s) | LSI_TAG_VALID;
+            s->select_tag |= lsi_get_msgbyte(s) | LSI_TAG_VALID;
             break;
         case 0x22: /* ORDERED queue */
             BADF("ORDERED queue not implemented\n");
-            s->current_tag |= lsi_get_msgbyte(s) | LSI_TAG_VALID;
+            s->select_tag |= lsi_get_msgbyte(s) | LSI_TAG_VALID;
             break;
         default:
             if ((msg & 0x80) == 0) {
@@ -905,17 +899,17 @@ static void lsi_memcpy(LSIState *s, uint32_t dest, uint32_t src, int count)
 
 static void lsi_wait_reselect(LSIState *s)
 {
-    int i;
+    lsi_request *p;
+
     DPRINTF("Wait Reselect\n");
-    if (s->current_dma_len)
+
-        BADF("Reselect with pending DMA\n");
+    QTAILQ_FOREACH(p, &s->queue, next) {
-    for (i = 0; i < s->active_commands; i++) {
+        if (p->pending) {
-        if (s->queue[i].pending) {
+            lsi_reselect(s, p);
-            lsi_reselect(s, s->queue[i].tag);
             break;
         }
     }
-    if (s->current_dma_len == 0) {
+    if (s->current == NULL) {
         s->waiting = 1;
     }
 }
@@ -1093,8 +1087,8 @@ again:
                 /* ??? Linux drivers compain when this is set.  Maybe
                    it only applies in low-level mode (unimplemented).
                 lsi_script_scsi_interrupt(s, LSI_SIST0_CMP, 0); */
-                s->current_dev = s->bus.devs[id];
+                s->select_dev = s->bus.devs[id];
-                s->current_tag = id << 8;
+                s->select_tag = id << 8;
                 s->scntl1 |= LSI_SCNTL1_CON;
                 if (insn & (1 << 3)) {
                     s->socl |= LSI_SOCL_ATN;
@@ -2006,9 +2000,11 @@ static void lsi_pre_save(void *opaque)
 {
     LSIState *s = opaque;
 
-    assert(s->dma_buf == NULL);
+    if (s->current) {
-    assert(s->current_dma_len == 0);
+        assert(s->current->dma_buf == NULL);
-    assert(s->active_commands == 0);
+        assert(s->current->dma_len == 0);
+    }
+    assert(QTAILQ_EMPTY(&s->queue));
 }
 
 static const VMStateDescription vmstate_lsi_scsi = {
@@ -2101,8 +2097,6 @@ static int lsi_scsi_uninit(PCIDevice *d)
     cpu_unregister_io_memory(s->mmio_io_addr);
     cpu_unregister_io_memory(s->ram_io_addr);
 
-    qemu_free(s->queue);
-
     return 0;
 }
 
@@ -2138,9 +2132,7 @@ static int lsi_scsi_init(PCIDevice *dev)
                            PCI_BASE_ADDRESS_SPACE_MEMORY, lsi_mmio_mapfunc);
     pci_register_bar((struct PCIDevice *)s, 2, 0x2000,
                            PCI_BASE_ADDRESS_SPACE_MEMORY, lsi_ram_mapfunc);
-    s->queue = qemu_malloc(sizeof(lsi_queue));
+    QTAILQ_INIT(&s->queue);
-    s->queue_len = 1;
-    s->active_commands = 0;
 
     lsi_soft_reset(s);
 

hw/mc146818rtc.c

@@ -30,6 +30,8 @@
 
 //#define DEBUG_CMOS
 
+#define RTC_REINJECT_ON_ACK_COUNT 20
+
 #define RTC_SECONDS             0
 #define RTC_SECONDS_ALARM       1
 #define RTC_MINUTES             2
@@ -76,6 +78,7 @@ struct RTCState {
     int64_t next_periodic_time;
     /* second update */
     int64_t next_second_time;
+    uint16_t irq_reinject_on_ack_count;
     uint32_t irq_coalesced;
     uint32_t period;
     QEMUTimer *coalesced_timer;
@@ -180,6 +183,8 @@ static void rtc_periodic_timer(void *opaque)
         s->cmos_data[RTC_REG_C] |= 0xc0;
 #ifdef TARGET_I386
         if(rtc_td_hack) {
+            if (s->irq_reinject_on_ack_count >= RTC_REINJECT_ON_ACK_COUNT)
+                s->irq_reinject_on_ack_count = 0;		
             apic_reset_irq_delivered();
             rtc_irq_raise(s->irq);
             if (!apic_get_irq_delivered()) {
@@ -458,6 +463,18 @@ static uint32_t cmos_ioport_read(void *opaque, uint32_t addr)
         case RTC_REG_C:
             ret = s->cmos_data[s->cmos_index];
             qemu_irq_lower(s->irq);
+#ifdef TARGET_I386
+            if(s->irq_coalesced &&
+                    s->irq_reinject_on_ack_count < RTC_REINJECT_ON_ACK_COUNT) {
+                s->irq_reinject_on_ack_count++;
+                apic_reset_irq_delivered();
+                qemu_irq_raise(s->irq);
+                if (apic_get_irq_delivered())
+                    s->irq_coalesced--;
+                break;
+            }
+#endif
+
             s->cmos_data[RTC_REG_C] = 0x00;
             break;
         default:

hw/msix.c

@@ -14,12 +14,15 @@
 #include "hw.h"
 #include "msix.h"
 #include "pci.h"
+#define QEMU_KVM_NO_CPU
+#include "qemu-kvm.h"
 
 /* Declaration from linux/pci_regs.h */
 #define  PCI_CAP_ID_MSIX 0x11 /* MSI-X */
 #define  PCI_MSIX_FLAGS 2     /* Table at lower 11 bits */
 #define  PCI_MSIX_FLAGS_QSIZE	0x7FF
 #define  PCI_MSIX_FLAGS_ENABLE	(1 << 15)
+#define  PCI_MSIX_FLAGS_MASKALL	(1 << 14)
 #define  PCI_MSIX_FLAGS_BIRMASK	(7 << 0)
 
 /* MSI-X capability structure */
@@ -27,9 +30,10 @@
 #define MSIX_PBA_OFFSET 8
 #define MSIX_CAP_LENGTH 12
 
-/* MSI enable bit is in byte 1 in FLAGS register */
+/* MSI enable bit and maskall bit are in byte 1 in FLAGS register */
-#define MSIX_ENABLE_OFFSET (PCI_MSIX_FLAGS + 1)
+#define MSIX_CONTROL_OFFSET (PCI_MSIX_FLAGS + 1)
 #define MSIX_ENABLE_MASK (PCI_MSIX_FLAGS_ENABLE >> 8)
+#define MSIX_MASKALL_MASK (PCI_MSIX_FLAGS_MASKALL >> 8)
 
 /* MSI-X table format */
 #define MSIX_MSG_ADDR 0
@@ -60,6 +64,117 @@
 /* Flag for interrupt controller to declare MSI-X support */
 int msix_supported;
 
+#ifdef CONFIG_KVM
+/* KVM specific MSIX helpers */
+static void kvm_msix_free(PCIDevice *dev)
+{
+    int vector, changed = 0;
+    for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
+        if (dev->msix_entry_used[vector]) {
+            kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
+            changed = 1;
+        }
+    }
+    if (changed) {
+        kvm_commit_irq_routes(kvm_context);
+    }
+}
+
+static void kvm_msix_routing_entry(PCIDevice *dev, unsigned vector,
+                                   struct kvm_irq_routing_entry *entry)
+{
+    uint8_t *table_entry = dev->msix_table_page + vector * MSIX_ENTRY_SIZE;
+    entry->type = KVM_IRQ_ROUTING_MSI;
+    entry->flags = 0;
+    entry->u.msi.address_lo = pci_get_long(table_entry + MSIX_MSG_ADDR);
+    entry->u.msi.address_hi = pci_get_long(table_entry + MSIX_MSG_UPPER_ADDR);
+    entry->u.msi.data = pci_get_long(table_entry + MSIX_MSG_DATA);
+}
+
+static void kvm_msix_update(PCIDevice *dev, int vector,
+                            int was_masked, int is_masked)
+{
+    struct kvm_irq_routing_entry e = {}, *entry;
+    int mask_cleared = was_masked && !is_masked;
+    /* It is only legal to change an entry when it is masked. Therefore, it is
+     * enough to update the routing in kernel when mask is being cleared. */
+    if (!mask_cleared) {
+        return;
+    }
+    if (!dev->msix_entry_used[vector]) {
+        return;
+    }
+    entry = dev->msix_irq_entries + vector;
+    e.gsi = entry->gsi;
+    kvm_msix_routing_entry(dev, vector, &e);
+    if (memcmp(&entry->u.msi, &e.u.msi, sizeof entry->u.msi)) {
+        int r;
+        r = kvm_update_routing_entry(kvm_context, entry, &e);
+        if (r) {
+            fprintf(stderr, "%s: kvm_update_routing_entry failed: %s\n", __func__,
+		    strerror(-r));
+            exit(1);
+        }
+        memcpy(&entry->u.msi, &e.u.msi, sizeof entry->u.msi);
+        r = kvm_commit_irq_routes(kvm_context);
+        if (r) {
+            fprintf(stderr, "%s: kvm_commit_irq_routes failed: %s\n", __func__,
+		    strerror(-r));
+            exit(1);
+        }
+    }
+}
+
+static int kvm_msix_add(PCIDevice *dev, unsigned vector)
+{
+    struct kvm_irq_routing_entry *entry = dev->msix_irq_entries + vector;
+    int r;
+
+    if (!kvm_has_gsi_routing(kvm_context)) {
+        fprintf(stderr, "Warning: no MSI-X support found. "
+                "At least kernel 2.6.30 is required for MSI-X support.\n"
+               );
+        return -EOPNOTSUPP;
+    }
+
+    r = kvm_get_irq_route_gsi(kvm_context);
+    if (r < 0) {
+        fprintf(stderr, "%s: kvm_get_irq_route_gsi failed: %s\n", __func__, strerror(-r));
+        return r;
+    }
+    entry->gsi = r;
+    kvm_msix_routing_entry(dev, vector, entry);
+    r = kvm_add_routing_entry(kvm_context, entry);
+    if (r < 0) {
+        fprintf(stderr, "%s: kvm_add_routing_entry failed: %s\n", __func__, strerror(-r));
+        return r;
+    }
+
+    r = kvm_commit_irq_routes(kvm_context);
+    if (r < 0) {
+        fprintf(stderr, "%s: kvm_commit_irq_routes failed: %s\n", __func__, strerror(-r));
+        return r;
+    }
+    return 0;
+}
+
+static void kvm_msix_del(PCIDevice *dev, unsigned vector)
+{
+    if (dev->msix_entry_used[vector]) {
+        return;
+    }
+    kvm_del_routing_entry(kvm_context, &dev->msix_irq_entries[vector]);
+    kvm_commit_irq_routes(kvm_context);
+}
+#else
+
+static void kvm_msix_free(PCIDevice *dev) {}
+static void kvm_msix_update(PCIDevice *dev, int vector,
+                            int was_masked, int is_masked) {}
+static int kvm_msix_add(PCIDevice *dev, unsigned vector) { return -1; }
+static void kvm_msix_del(PCIDevice *dev, unsigned vector) {}
+#endif
+
 /* Add MSI-X capability to the config space for the device. */
 /* Given a bar and its size, add MSI-X table on top of it
  * and fill MSI-X capability in the config space.
@@ -101,22 +216,11 @@ static int msix_add_config(struct PCIDevice *pdev, unsigned short nentries,
                  bar_nr);
     pdev->msix_cap = config_offset;
     /* Make flags bit writeable. */
-    pdev->wmask[config_offset + MSIX_ENABLE_OFFSET] |= MSIX_ENABLE_MASK;
+    pdev->wmask[config_offset + MSIX_CONTROL_OFFSET] |= MSIX_ENABLE_MASK |
+	    MSIX_MASKALL_MASK;
     return 0;
 }
 
-/* Handle MSI-X capability config write. */
-void msix_write_config(PCIDevice *dev, uint32_t addr,
-                       uint32_t val, int len)
-{
-    unsigned enable_pos = dev->msix_cap + MSIX_ENABLE_OFFSET;
-    if (addr + len <= enable_pos || addr > enable_pos)
-        return;
-
-    if (msix_enabled(dev))
-        qemu_set_irq(dev->irq[0], 0);
-}
-
 static uint32_t msix_mmio_readl(void *opaque, target_phys_addr_t addr)
 {
     PCIDevice *dev = opaque;
@@ -157,10 +261,50 @@ static void msix_clr_pending(PCIDevice *dev, int vector)
     *msix_pending_byte(dev, vector) &= ~msix_pending_mask(vector);
 }
 
+static int msix_function_masked(PCIDevice *dev)
+{
+    return dev->config[dev->msix_cap + MSIX_CONTROL_OFFSET] & MSIX_MASKALL_MASK;
+}
+
 static int msix_is_masked(PCIDevice *dev, int vector)
 {
     unsigned offset = vector * MSIX_ENTRY_SIZE + MSIX_VECTOR_CTRL;
-    return dev->msix_table_page[offset] & MSIX_VECTOR_MASK;
+    return msix_function_masked(dev) ||
+	   dev->msix_table_page[offset] & MSIX_VECTOR_MASK;
+}
+
+static void msix_handle_mask_update(PCIDevice *dev, int vector)
+{
+    if (!msix_is_masked(dev, vector) && msix_is_pending(dev, vector)) {
+        msix_clr_pending(dev, vector);
+        msix_notify(dev, vector);
+    }
+}
+
+/* Handle MSI-X capability config write. */
+void msix_write_config(PCIDevice *dev, uint32_t addr,
+                       uint32_t val, int len)
+{
+    unsigned enable_pos = dev->msix_cap + MSIX_CONTROL_OFFSET;
+    int vector;
+
+    if (addr + len <= enable_pos || addr > enable_pos) {
+        return;
+    }
+
+    if (!msix_enabled(dev)) {
+        return;
+    }
+
+    qemu_set_irq(dev->irq[0], 0);
+
+    if (msix_function_masked(dev)) {
+        return;
+    }
+
+    for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
+        msix_handle_mask_update(dev, vector);
+    }
 }
 
 static void msix_mmio_writel(void *opaque, target_phys_addr_t addr,
@@ -169,11 +313,12 @@ static void msix_mmio_writel(void *opaque, target_phys_addr_t addr,
     PCIDevice *dev = opaque;
     unsigned int offset = addr & (MSIX_PAGE_SIZE - 1) & ~0x3;
     int vector = offset / MSIX_ENTRY_SIZE;
+    int was_masked = msix_is_masked(dev, vector);
     pci_set_long(dev->msix_table_page + offset, val);
-    if (!msix_is_masked(dev, vector) && msix_is_pending(dev, vector)) {
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
-        msix_clr_pending(dev, vector);
+        kvm_msix_update(dev, vector, was_masked, msix_is_masked(dev, vector));
-        msix_notify(dev, vector);
     }
+    msix_handle_mask_update(dev, vector);
 }
 
 static void msix_mmio_write_unallowed(void *opaque, target_phys_addr_t addr,
@@ -231,6 +376,12 @@ int msix_init(struct PCIDevice *dev, unsigned short nentries,
     if (nentries > MSIX_MAX_ENTRIES)
         return -EINVAL;
 
+#ifdef KVM_CAP_IRQCHIP
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        dev->msix_irq_entries = qemu_malloc(nentries *
+                                            sizeof *dev->msix_irq_entries);
+    }
+#endif
     dev->msix_entry_used = qemu_mallocz(MSIX_MAX_ENTRIES *
                                         sizeof *dev->msix_entry_used);
 
@@ -267,6 +418,10 @@ static void msix_free_irq_entries(PCIDevice *dev)
 {
     int vector;
 
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_msix_free(dev);
+    }
+
     for (vector = 0; vector < dev->msix_entries_nr; ++vector) {
         dev->msix_entry_used[vector] = 0;
         msix_clr_pending(dev, vector);
@@ -287,6 +442,8 @@ int msix_uninit(PCIDevice *dev)
     dev->msix_table_page = NULL;
     qemu_free(dev->msix_entry_used);
     dev->msix_entry_used = NULL;
+    qemu_free(dev->msix_irq_entries);
+    dev->msix_irq_entries = NULL;
     dev->cap_present &= ~QEMU_PCI_CAP_MSIX;
     return 0;
 }
@@ -295,10 +452,13 @@ void msix_save(PCIDevice *dev, QEMUFile *f)
 {
     unsigned n = dev->msix_entries_nr;
 
-    if (!(dev->cap_present & QEMU_PCI_CAP_MSIX)) {
+    if (!msix_supported) {
         return;
     }
 
+    if (!(dev->cap_present & QEMU_PCI_CAP_MSIX)) {
+        return;
+    }
     qemu_put_buffer(f, dev->msix_table_page, n * MSIX_ENTRY_SIZE);
     qemu_put_buffer(f, dev->msix_table_page + MSIX_PAGE_PENDING, (n + 7) / 8);
 }
@@ -308,6 +468,9 @@ void msix_load(PCIDevice *dev, QEMUFile *f)
 {
     unsigned n = dev->msix_entries_nr;
 
+    if (!msix_supported)
+        return;
+
     if (!(dev->cap_present & QEMU_PCI_CAP_MSIX)) {
         return;
     }
@@ -327,7 +490,7 @@ int msix_present(PCIDevice *dev)
 int msix_enabled(PCIDevice *dev)
 {
     return (dev->cap_present & QEMU_PCI_CAP_MSIX) &&
-        (dev->config[dev->msix_cap + MSIX_ENABLE_OFFSET] &
+        (dev->config[dev->msix_cap + MSIX_CONTROL_OFFSET] &
          MSIX_ENABLE_MASK);
 }
 
@@ -352,6 +515,13 @@ void msix_notify(PCIDevice *dev, unsigned vector)
         return;
     }
 
+#ifdef KVM_CAP_IRQCHIP
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_set_irq(dev->msix_irq_entries[vector].gsi, 1, NULL);
+        return;
+    }
+#endif
+
     address = pci_get_long(table_entry + MSIX_MSG_UPPER_ADDR);
     address = (address << 32) | pci_get_long(table_entry + MSIX_MSG_ADDR);
     data = pci_get_long(table_entry + MSIX_MSG_DATA);
@@ -363,8 +533,8 @@ void msix_reset(PCIDevice *dev)
     if (!(dev->cap_present & QEMU_PCI_CAP_MSIX))
         return;
     msix_free_irq_entries(dev);
-    dev->config[dev->msix_cap + MSIX_ENABLE_OFFSET] &=
+    dev->config[dev->msix_cap + MSIX_CONTROL_OFFSET] &=
-	    ~dev->wmask[dev->msix_cap + MSIX_ENABLE_OFFSET];
+	    ~dev->wmask[dev->msix_cap + MSIX_CONTROL_OFFSET];
     memset(dev->msix_table_page, 0, MSIX_PAGE_SIZE);
     msix_mask_all(dev, dev->msix_entries_nr);
 }
@@ -380,9 +550,19 @@ void msix_reset(PCIDevice *dev)
 /* Mark vector as used. */
 int msix_vector_use(PCIDevice *dev, unsigned vector)
 {
+    int ret;
     if (vector >= dev->msix_entries_nr)
         return -EINVAL;
-    dev->msix_entry_used[vector]++;
+    if (dev->msix_entry_used[vector]) {
+        return 0;
+    }
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        ret = kvm_msix_add(dev, vector);
+        if (ret) {
+            return ret;
+        }
+    }
+    ++dev->msix_entry_used[vector];
     return 0;
 }
 
@@ -395,6 +575,9 @@ void msix_vector_unuse(PCIDevice *dev, unsigned vector)
     if (--dev->msix_entry_used[vector]) {
         return;
     }
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
+        kvm_msix_del(dev, vector);
+    }
     msix_clr_pending(dev, vector);
 }
 

hw/musicpal.c

@@ -67,7 +67,7 @@
 #define MP_AUDIO_IRQ            30
 
 /* Wolfson 8750 I2C address */
-#define MP_WM_ADDR              0x34
+#define MP_WM_ADDR              0x1A
 
 /* Ethernet register offsets */
 #define MP_ETH_SMIR             0x010
@@ -238,14 +238,13 @@ static void eth_send(mv88w8618_eth_state *s, int queue_index)
 {
     uint32_t desc_addr = s->tx_queue[queue_index];
     mv88w8618_tx_desc desc;
+    uint32_t next_desc;
     uint8_t buf[2048];
     int len;
 
-    if (!desc_addr) {
-        return;
-    }
     do {
         eth_tx_desc_get(desc_addr, &desc);
+        next_desc = desc.next;
         if (desc.cmdstat & MP_ETH_TX_OWN) {
             len = desc.bytes;
             if (len < 2048) {
@@ -256,7 +255,7 @@ static void eth_send(mv88w8618_eth_state *s, int queue_index)
             s->icr |= 1 << (MP_ETH_IRQ_TXLO_BIT - queue_index);
             eth_tx_desc_put(desc_addr, &desc);
         }
-        desc_addr = desc.next;
+        desc_addr = next_desc;
     } while (desc_addr != s->tx_queue[queue_index]);
 }
 

hw/pc.c

@@ -44,6 +44,9 @@
 #include "ide.h"
 #include "loader.h"
 #include "elf.h"
+#include "device-assignment.h"
+
+#include "qemu-kvm.h"
 
 /* output Bochs bios info messages */
 //#define DEBUG_BIOS
@@ -52,6 +55,8 @@
 //#define DEBUG_MULTIBOOT
 
 #define BIOS_FILENAME "bios.bin"
+#define EXTBOOT_FILENAME "extboot.bin"
+#define VAPIC_FILENAME "vapic.bin"
 
 #define PC_MAX_BIOS_SIZE (4 * 1024 * 1024)
 
@@ -69,6 +74,8 @@ static RTCState *rtc_state;
 static PITState *pit;
 static PCII440FXState *i440fx_state;
 
+qemu_irq *ioapic_irq_hack;
+
 typedef struct isa_irq_state {
     qemu_irq *i8259;
     qemu_irq *ioapic;
@@ -560,19 +567,21 @@ static int load_multiboot(void *fw_cfg,
     }
     if (!(flags & 0x00010000)) { /* MULTIBOOT_HEADER_HAS_ADDR */
         uint64_t elf_entry;
+        uint64_t elf_low, elf_high;
         int kernel_size;
         fclose(f);
-        kernel_size = load_elf(kernel_filename, 0, &elf_entry, NULL, NULL,
+        kernel_size = load_elf(kernel_filename, 0, &elf_entry, &elf_low, &elf_high,
                                0, ELF_MACHINE, 0);
         if (kernel_size < 0) {
             fprintf(stderr, "Error while loading elf kernel\n");
             exit(1);
         }
-        mh_load_addr = mh_entry_addr = elf_entry;
+        mh_load_addr = elf_low;
-        mb_kernel_size = kernel_size;
+        mb_kernel_size = elf_high - elf_low;
+        mh_entry_addr = elf_entry;
 
         mb_kernel_data = qemu_malloc(mb_kernel_size);
-        if (rom_copy(mb_kernel_data, elf_entry, kernel_size) != kernel_size) {
+        if (rom_copy(mb_kernel_data, mh_load_addr, mb_kernel_size) != mb_kernel_size) {
             fprintf(stderr, "Error while fetching elf kernel from rom\n");
             exit(1);
         }
@@ -950,7 +959,7 @@ int cpu_is_bsp(CPUState *env)
     return env->cpuid_apic_id == 0;
 }
 
-static CPUState *pc_new_cpu(const char *cpu_model)
+CPUState *pc_new_cpu(const char *cpu_model)
 {
     CPUState *env;
 
@@ -959,6 +968,7 @@ static CPUState *pc_new_cpu(const char *cpu_model)
         fprintf(stderr, "Unable to find x86 CPU definition\n");
         exit(1);
     }
+    env->kvm_cpu_state.regs_modified = 1;
     if ((env->cpuid_features & CPUID_APIC) || smp_cpus > 1) {
         env->cpuid_apic_id = env->cpu_index;
         /* APIC reset callback resets cpu */
@@ -966,6 +976,11 @@ static CPUState *pc_new_cpu(const char *cpu_model)
     } else {
         qemu_register_reset((QEMUResetHandler*)cpu_reset, env);
     }
+
+    /* kvm needs this to run after the apic is initialized. Otherwise,
+     * it can access invalid state and crash.
+     */
+    qemu_init_vcpu(env);
     return env;
 }
 
@@ -1013,6 +1028,9 @@ static void pc_init1(ram_addr_t ram_size,
 #endif
     }
 
+    if (kvm_enabled()) {
+        kvm_set_boot_cpu_id(0);
+    }
     for (i = 0; i < smp_cpus; i++) {
         env = pc_new_cpu(cpu_model);
     }
@@ -1020,18 +1038,11 @@ static void pc_init1(ram_addr_t ram_size,
     vmport_init();
 
     /* allocate RAM */
-    ram_addr = qemu_ram_alloc(0xa0000);
+    ram_addr = qemu_ram_alloc(below_4g_mem_size);
     cpu_register_physical_memory(0, 0xa0000, ram_addr);
-
-    /* Allocate, even though we won't register, so we don't break the
-     * phys_ram_base + PA assumption. This range includes vga (0xa0000 - 0xc0000),
-     * and some bios areas, which will be registered later
-     */
-    ram_addr = qemu_ram_alloc(0x100000 - 0xa0000);
-    ram_addr = qemu_ram_alloc(below_4g_mem_size - 0x100000);
     cpu_register_physical_memory(0x100000,
                  below_4g_mem_size - 0x100000,
-                 ram_addr);
+                 ram_addr + 0x100000);
 
     /* above 4giga memory allocation */
     if (above_4g_mem_size > 0) {
@@ -1073,11 +1084,17 @@ static void pc_init1(ram_addr_t ram_size,
     isa_bios_size = bios_size;
     if (isa_bios_size > (128 * 1024))
         isa_bios_size = 128 * 1024;
+    cpu_register_physical_memory(0xd0000, (192 * 1024) - isa_bios_size,
+                                 IO_MEM_UNASSIGNED);
+    /* kvm tpr optimization needs the bios accessible for write, at least to qemu itself */
     cpu_register_physical_memory(0x100000 - isa_bios_size,
                                  isa_bios_size,
-                                 (bios_offset + bios_size - isa_bios_size) | IO_MEM_ROM);
+                                 (bios_offset + bios_size - isa_bios_size) /* | IO_MEM_ROM */);
-
 
+    if (extboot_drive) {
+        option_rom[nb_option_roms++] = qemu_strdup(EXTBOOT_FILENAME);
+    }
+    option_rom[nb_option_roms++] = qemu_strdup(VAPIC_FILENAME);
 
     rom_enable_driver_roms = 1;
     option_rom_offset = qemu_ram_alloc(PC_ROM_SIZE);
@@ -1088,6 +1105,7 @@ static void pc_init1(ram_addr_t ram_size,
                                  bios_size, bios_offset | IO_MEM_ROM);
 
     fw_cfg = bochs_bios_init();
+    rom_set_fw(fw_cfg);
 
     if (linux_boot) {
         load_linux(fw_cfg, kernel_filename, initrd_filename, kernel_cmdline, below_4g_mem_size);
@@ -1098,10 +1116,18 @@ static void pc_init1(ram_addr_t ram_size,
     }
 
     cpu_irq = qemu_allocate_irqs(pic_irq_request, NULL, 1);
-    i8259 = i8259_init(cpu_irq[0]);
+#ifdef KVM_CAP_IRQCHIP
-    isa_irq_state = qemu_mallocz(sizeof(*isa_irq_state));
+    if (kvm_enabled() && kvm_irqchip_in_kernel()) {
-    isa_irq_state->i8259 = i8259;
+        isa_irq_state = qemu_mallocz(sizeof(*isa_irq_state));
-    isa_irq = qemu_allocate_irqs(isa_irq_handler, isa_irq_state, 24);
+        isa_irq = i8259 = kvm_i8259_init(cpu_irq[0]);
+    } else
+#endif
+    {
+        i8259 = i8259_init(cpu_irq[0]);
+        isa_irq_state = qemu_mallocz(sizeof(*isa_irq_state));
+        isa_irq_state->i8259 = i8259;
+        isa_irq = qemu_allocate_irqs(isa_irq_handler, isa_irq_state, 24);
+    }
 
     if (pci_enabled) {
         pci_bus = i440fx_init(&i440fx_state, &piix3_devfn, isa_irq);
@@ -1146,8 +1172,14 @@ static void pc_init1(ram_addr_t ram_size,
 
     if (pci_enabled) {
         isa_irq_state->ioapic = ioapic_init();
+        ioapic_irq_hack = isa_irq;
     }
-    pit = pit_init(0x40, isa_reserve_irq(0));
+#ifdef CONFIG_KVM_PIT
+    if (kvm_enabled() && qemu_kvm_pit_in_kernel())
+	pit = kvm_pit_init(0x40, isa_reserve_irq(0));
+    else
+#endif
+	pit = pit_init(0x40, isa_reserve_irq(0));
     pcspk_init(pit);
     if (!no_hpet) {
         hpet_init(isa_irq);
@@ -1171,7 +1203,7 @@ static void pc_init1(ram_addr_t ram_size,
         if (!pci_enabled || (nd->model && strcmp(nd->model, "ne2k_isa") == 0))
             pc_init_ne2k_isa(nd);
         else
-            pci_nic_init_nofail(nd, "e1000", NULL);
+            pci_nic_init_nofail(nd, "rtl8139", NULL);
     }
 
     if (drive_get_max_bus(IF_IDE) >= MAX_IDE_BUS) {
@@ -1223,7 +1255,7 @@ static void pc_init1(ram_addr_t ram_size,
             qdev_prop_set_ptr(eeprom, "data", eeprom_buf + (i * 256));
             qdev_init_nofail(eeprom);
         }
-        piix4_acpi_system_hot_add_init(pci_bus);
+        piix4_acpi_system_hot_add_init(pci_bus, cpu_model);
     }
 
     if (i440fx_state) {
@@ -1240,6 +1272,18 @@ static void pc_init1(ram_addr_t ram_size,
         }
     }
 
+    if (extboot_drive) {
+	DriveInfo *info = extboot_drive;
+	int cyls, heads, secs;
+
+	if (info->type != IF_IDE && info->type != IF_VIRTIO) {
+	    bdrv_guess_geometry(info->bdrv, &cyls, &heads, &secs);
+	    bdrv_set_geometry_hint(info->bdrv, cyls, heads, secs);
+	}
+
+	extboot_init(info->bdrv, 1);
+    }
+
     /* Add virtio console devices */
     if (pci_enabled) {
         for(i = 0; i < MAX_VIRTIO_CONSOLES; i++) {
@@ -1248,6 +1292,12 @@ static void pc_init1(ram_addr_t ram_size,
             }
         }
     }
+
+#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
+    if (kvm_enabled()) {
+        add_assigned_devices(pci_bus, assigned_devices, assigned_devices_index);
+    }
+#endif /* CONFIG_KVM_DEVICE_ASSIGNMENT */
 }
 
 static void pc_init_pci(ram_addr_t ram_size,
@@ -1285,7 +1335,7 @@ void cmos_set_s3_resume(void)
 }
 
 static QEMUMachine pc_machine = {
-    .name = "pc-0.11",
+    .name = "pc-0.12",
     .alias = "pc",
     .desc = "Standard PC",
     .init = pc_init_pci,
@@ -1293,12 +1343,39 @@ static QEMUMachine pc_machine = {
     .is_default = 1,
 };
 
+static QEMUMachine pc_machine_v0_11 = {
+    .name = "pc-0.11",
+    .desc = "Standard PC, qemu 0.11",
+    .init = pc_init_pci,
+    .max_cpus = 255,
+    .compat_props = (GlobalProperty[]) {
+        {
+            .driver   = "virtio-blk-pci",
+            .property = "vectors",
+            .value    = stringify(0),
+        },{
+            .driver   = "ide-drive",
+            .property = "ver",
+            .value    = "0.11",
+        },{
+            .driver   = "scsi-disk",
+            .property = "ver",
+            .value    = "0.11",
+        },{
+            .driver   = "PCI",
+            .property = "rombar",
+            .value    = stringify(0),
+        },
+        { /* end of list */ }
+    }
+};
+
 static QEMUMachine pc_machine_v0_10 = {
     .name = "pc-0.10",
     .desc = "Standard PC, qemu 0.10",
     .init = pc_init_pci,
     .max_cpus = 255,
-    .compat_props = (CompatProperty[]) {
+    .compat_props = (GlobalProperty[]) {
         {
             .driver   = "virtio-blk-pci",
             .property = "class",
@@ -1315,6 +1392,18 @@ static QEMUMachine pc_machine_v0_10 = {
             .driver   = "virtio-blk-pci",
             .property = "vectors",
             .value    = stringify(0),
+        },{
+            .driver   = "ide-drive",
+            .property = "ver",
+            .value    = "0.10",
+        },{
+            .driver   = "scsi-disk",
+            .property = "ver",
+            .value    = "0.10",
+        },{
+            .driver   = "PCI",
+            .property = "rombar",
+            .value    = stringify(0),
         },
         { /* end of list */ }
     },
@@ -1330,6 +1419,7 @@ static QEMUMachine isapc_machine = {
 static void pc_machine_init(void)
 {
     qemu_register_machine(&pc_machine);
+    qemu_register_machine(&pc_machine_v0_11);
     qemu_register_machine(&pc_machine_v0_10);
     qemu_register_machine(&isapc_machine);
 }

hw/pc.h

@@ -28,6 +28,7 @@ extern PicState2 *isa_pic;
 void pic_set_irq(int irq, int level);
 void pic_set_irq_new(void *opaque, int irq, int level);
 qemu_irq *i8259_init(qemu_irq parent_irq);
+qemu_irq *kvm_i8259_init(qemu_irq parent_irq);
 int pic_read_irq(PicState2 *s);
 void pic_update_irq(PicState2 *s);
 uint32_t pic_intack_read(PicState2 *s);
@@ -48,6 +49,7 @@ qemu_irq *ioapic_init(void);
 void ioapic_set_irq(void *opaque, int vector, int level);
 void apic_reset_irq_delivered(void);
 int apic_get_irq_delivered(void);
+void apic_set_irq_delivered(void);
 
 /* i8254.c */
 
@@ -62,8 +64,12 @@ int pit_get_initial_count(PITState *pit, int channel);
 int pit_get_mode(PITState *pit, int channel);
 int pit_get_out(PITState *pit, int channel, int64_t current_time);
 
-void hpet_pit_disable(void);
+/* i8254-kvm.c */
-void hpet_pit_enable(void);
+
+PITState *kvm_pit_init(int base, qemu_irq irq);
+
+void hpet_disable_pit(void);
+void hpet_enable_pit(void);
 
 /* vmport.c */
 void vmport_init(void);
@@ -93,6 +99,7 @@ extern int fd_bootchk;
 
 void ioport_set_a20(int enable);
 int ioport_get_a20(void);
+CPUState *pc_new_cpu(const char *cpu_model);
 
 /* acpi.c */
 extern int acpi_enabled;
@@ -106,7 +113,7 @@ int acpi_table_add(const char *table_desc);
 i2c_bus *piix4_pm_init(PCIBus *bus, int devfn, uint32_t smb_io_base,
                        qemu_irq sci_irq);
 void piix4_smbus_register_device(SMBusDevice *dev, uint8_t addr);
-void piix4_acpi_system_hot_add_init(PCIBus *bus);
+void piix4_acpi_system_hot_add_init(PCIBus *bus, const char *model);
 
 /* hpet.c */
 extern int no_hpet;
@@ -116,6 +123,9 @@ void pcspk_init(PITState *);
 int pcspk_audio_init(qemu_irq *pic);
 
 /* piix_pci.c */
+/* config space register for IRQ routing */
+#define PIIX_CONFIG_IRQ_ROUTE 0x60
+
 struct PCII440FXState;
 typedef struct PCII440FXState PCII440FXState;
 
@@ -127,6 +137,10 @@ void i440fx_init_memory_mappings(PCII440FXState *d);
 extern PCIDevice *piix4_dev;
 int piix4_init(PCIBus *bus, int devfn);
 
+int piix_get_irq(int pin);
+
+int ipf_map_irq(PCIDevice *pci_dev, int irq_num);
+
 /* vga.c */
 enum vga_retrace_method {
     VGA_RETRACE_DUMB,
@@ -149,5 +163,10 @@ void isa_cirrus_vga_init(void);
 
 void isa_ne2000_init(int base, int irq, NICInfo *nd);
 
+/* extboot.c */
+
+void extboot_init(BlockDriverState *bs, int cmd);
+
 int cpu_is_bsp(CPUState *env);
+
 #endif

hw/pci-hotplug.c

@@ -33,6 +33,8 @@
 #include "scsi.h"
 #include "virtio-blk.h"
 #include "qemu-config.h"
+#include "qemu-objects.h"
+#include "device-assignment.h"
 
 #if defined(TARGET_I386)
 static PCIDevice *qemu_pci_hot_add_nic(Monitor *mon,
@@ -40,7 +42,18 @@ static PCIDevice *qemu_pci_hot_add_nic(Monitor *mon,
                                        const char *opts_str)
 {
     QemuOpts *opts;
-    int ret;
+    PCIBus *bus;
+    int ret, devfn;
+
+    bus = pci_get_bus_devfn(&devfn, devaddr);
+    if (!bus) {
+        monitor_printf(mon, "Invalid PCI device address %s\n", devaddr);
+        return NULL;
+    }
+    if (!((BusState*)bus)->allow_hotplug) {
+        monitor_printf(mon, "PCI bus doesn't support hotplug\n");
+        return NULL;
+    }
 
     opts = qemu_opts_parse(&qemu_net_opts, opts_str ? opts_str : "", NULL);
     if (!opts) {
@@ -82,6 +95,7 @@ static int scsi_hot_add(DeviceState *adapter, DriveInfo *dinfo, int printinfo)
      */
     dinfo->unit = qemu_opt_get_number(dinfo->opts, "unit", -1);
     scsidev = scsi_bus_legacy_add_drive(scsibus, dinfo, dinfo->unit);
+    dinfo->unit = scsidev->id;
 
     if (printinfo)
         qemu_error("OK bus %d, unit %d\n", scsibus->busnr, scsidev->id);
@@ -179,17 +193,17 @@ static PCIDevice *qemu_pci_hot_add_storage(Monitor *mon,
         monitor_printf(mon, "Invalid PCI device address %s\n", devaddr);
         return NULL;
     }
+    if (!((BusState*)bus)->allow_hotplug) {
+        monitor_printf(mon, "PCI bus doesn't support hotplug\n");
+        return NULL;
+    }
 
     switch (type) {
     case IF_SCSI:
-        if (!dinfo) {
-            monitor_printf(mon, "scsi requires a backing file/device.\n");
-            return NULL;
-        }
         dev = pci_create(bus, devfn, "lsi53c895a");
         if (qdev_init(&dev->qdev) < 0)
             dev = NULL;
-        if (dev) {
+        if (dev && dinfo) {
             if (scsi_hot_add(&dev->qdev, dinfo, 0) != 0) {
                 qdev_unplug(&dev->qdev);
                 dev = NULL;
@@ -212,7 +226,54 @@ static PCIDevice *qemu_pci_hot_add_storage(Monitor *mon,
     return dev;
 }
 
-void pci_device_hot_add(Monitor *mon, const QDict *qdict)
+#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
+static PCIDevice *qemu_pci_hot_assign_device(Monitor *mon,
+                                             const char *devaddr,
+                                             const char *opts_str)
+{
+    QemuOpts *opts;
+    DeviceState *dev;
+
+    opts = add_assigned_device(opts_str);
+    if (opts == NULL) {
+        monitor_printf(mon, "Error adding device; check syntax\n");
+        return NULL;
+    }
+    dev = qdev_device_add(opts);
+    return DO_UPCAST(PCIDevice, qdev, dev);
+}
+#endif /* CONFIG_KVM_DEVICE_ASSIGNMENT */
+
+void pci_device_hot_add_print(Monitor *mon, const QObject *data)
+{
+    QDict *qdict;
+
+    assert(qobject_type(data) == QTYPE_QDICT);
+    qdict = qobject_to_qdict(data);
+
+    monitor_printf(mon, "OK domain %d, bus %d, slot %d, function %d\n",
+                   (int) qdict_get_int(qdict, "domain"),
+                   (int) qdict_get_int(qdict, "bus"),
+                   (int) qdict_get_int(qdict, "slot"),
+                   (int) qdict_get_int(qdict, "function"));
+
+}
+
+/**
+ * pci_device_hot_add(): Hot add a PCI device
+ *
+ * Return a QDict with the following device information:
+ *
+ * - "domain": domain number
+ * - "bus": bus number
+ * - "slot": slot number
+ * - "function": function number
+ *
+ * Example:
+ *
+ * { "domain": 0, "bus": 0, "slot": 5, "function": 0 }
+ */
+void pci_device_hot_add(Monitor *mon, const QDict *qdict, QObject **ret_data)
 {
     PCIDevice *dev = NULL;
     const char *pci_addr = qdict_get_str(qdict, "pci_addr");
@@ -235,13 +296,19 @@ void pci_device_hot_add(Monitor *mon, const QDict *qdict)
         dev = qemu_pci_hot_add_nic(mon, pci_addr, opts);
     else if (strcmp(type, "storage") == 0)
         dev = qemu_pci_hot_add_storage(mon, pci_addr, opts);
+#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
+    else if (strcmp(type, "host") == 0)
+        dev = qemu_pci_hot_assign_device(mon, pci_addr, opts);
+#endif /* CONFIG_KVM_DEVICE_ASSIGNMENT */
     else
         monitor_printf(mon, "invalid type: %s\n", type);
 
     if (dev) {
-        monitor_printf(mon, "OK domain %d, bus %d, slot %d, function %d\n",
+        *ret_data =
-                       0, pci_bus_num(dev->bus), PCI_SLOT(dev->devfn),
+        qobject_from_jsonf("{ 'domain': 0, 'bus': %d, 'slot': %d, "
-                       PCI_FUNC(dev->devfn));
+                           "'function': %d }", pci_bus_num(dev->bus),
+                           PCI_SLOT(dev->devfn), PCI_FUNC(dev->devfn));
+        assert(*ret_data != NULL);
     } else
         monitor_printf(mon, "failed to add %s\n", opts);
 }

hw/pci.c

@@ -26,6 +26,10 @@
 #include "monitor.h"
 #include "net.h"
 #include "sysemu.h"
+#include "loader.h"
+#include "qemu-kvm.h"
+#include "hw/pc.h"
+#include "device-assignment.h"
 
 //#define DEBUG_PCI
 #ifdef DEBUG_PCI
@@ -62,12 +66,15 @@ static struct BusInfo pci_bus_info = {
     .print_dev  = pcibus_dev_print,
     .props      = (Property[]) {
         DEFINE_PROP_PCI_DEVFN("addr", PCIDevice, devfn, -1),
+        DEFINE_PROP_STRING("romfile", PCIDevice, romfile),
+        DEFINE_PROP_UINT32("rombar",  PCIDevice, rom_bar, 1),
         DEFINE_PROP_END_OF_LIST()
     }
 };
 
 static void pci_update_mappings(PCIDevice *d);
 static void pci_set_irq(void *opaque, int irq_num, int level);
+static int pci_add_option_rom(PCIDevice *pdev);
 
 target_phys_addr_t pci_mem_base;
 static uint16_t pci_default_sub_vendor_id = PCI_SUBVENDOR_ID_REDHAT_QUMRANET;
@@ -103,11 +110,48 @@ static int pci_bar(PCIDevice *d, int reg)
     return type == PCI_HEADER_TYPE_BRIDGE ? PCI_ROM_ADDRESS1 : PCI_ROM_ADDRESS;
 }
 
+static inline int pci_irq_state(PCIDevice *d, int irq_num)
+{
+	return (d->irq_state >> irq_num) & 0x1;
+}
+
+static inline void pci_set_irq_state(PCIDevice *d, int irq_num, int level)
+{
+	d->irq_state &= ~(0x1 << irq_num);
+	d->irq_state |= level << irq_num;
+}
+
+static void pci_change_irq_level(PCIDevice *pci_dev, int irq_num, int change)
+{
+    PCIBus *bus;
+    for (;;) {
+        bus = pci_dev->bus;
+        irq_num = bus->map_irq(pci_dev, irq_num);
+        if (bus->set_irq)
+            break;
+        pci_dev = bus->parent_dev;
+    }
+    bus->irq_count[irq_num] += change;
+    bus->set_irq(bus->irq_opaque, irq_num, bus->irq_count[irq_num] != 0);
+}
+
+/* Update interrupt status bit in config space on interrupt
+ * state change. */
+static void pci_update_irq_status(PCIDevice *dev)
+{
+    if (dev->irq_state) {
+        dev->config[PCI_STATUS] |= PCI_STATUS_INTERRUPT;
+    } else {
+        dev->config[PCI_STATUS] &= ~PCI_STATUS_INTERRUPT;
+    }
+}
+
 static void pci_device_reset(PCIDevice *dev)
 {
     int r;
 
-    memset(dev->irq_state, 0, sizeof dev->irq_state);
+    dev->irq_state = 0;
+    pci_update_irq_status(dev);
     dev->config[PCI_COMMAND] &= ~(PCI_COMMAND_IO | PCI_COMMAND_MEMORY |
                                   PCI_COMMAND_MASTER);
     dev->config[PCI_CACHE_LINE_SIZE] = 0x0;
@@ -274,6 +318,43 @@ static VMStateInfo vmstate_info_pci_config = {
     .put  = put_pci_config_device,
 };
 
+static int get_pci_irq_state(QEMUFile *f, void *pv, size_t size)
+{
+    PCIDevice *s = container_of(pv, PCIDevice, irq_state);
+    uint32_t irq_state[PCI_NUM_PINS];
+    int i;
+    for (i = 0; i < PCI_NUM_PINS; ++i) {
+        irq_state[i] = qemu_get_be32(f);
+        if (irq_state[i] != 0x1 && irq_state[i] != 0) {
+            fprintf(stderr, "irq state %d: must be 0 or 1.\n",
+                    irq_state[i]);
+            return -EINVAL;
+        }
+    }
+
+    for (i = 0; i < PCI_NUM_PINS; ++i) {
+        pci_set_irq_state(s, i, irq_state[i]);
+    }
+
+    return 0;
+}
+
+static void put_pci_irq_state(QEMUFile *f, void *pv, size_t size)
+{
+    int i;
+    PCIDevice *s = container_of(pv, PCIDevice, irq_state);
+
+    for (i = 0; i < PCI_NUM_PINS; ++i) {
+        qemu_put_be32(f, pci_irq_state(s, i));
+    }
+}
+
+static VMStateInfo vmstate_info_pci_irq_state = {
+    .name = "pci irq state",
+    .get  = get_pci_irq_state,
+    .put  = put_pci_irq_state,
+};
+
 const VMStateDescription vmstate_pci_device = {
     .name = "PCIDevice",
     .version_id = 2,
@@ -284,7 +365,9 @@ const VMStateDescription vmstate_pci_device = {
         VMSTATE_BUFFER_UNSAFE_INFO(config, PCIDevice, 0,
                                    vmstate_info_pci_config,
                                    PCI_CONFIG_SPACE_SIZE),
-        VMSTATE_INT32_ARRAY_V(irq_state, PCIDevice, PCI_NUM_PINS, 2),
+        VMSTATE_BUFFER_UNSAFE_INFO(irq_state, PCIDevice, 2,
+				   vmstate_info_pci_irq_state,
+				   PCI_NUM_PINS * sizeof(int32_t)),
         VMSTATE_END_OF_LIST()
     }
 };
@@ -299,7 +382,9 @@ const VMStateDescription vmstate_pcie_device = {
         VMSTATE_BUFFER_UNSAFE_INFO(config, PCIDevice, 0,
                                    vmstate_info_pci_config,
                                    PCIE_CONFIG_SPACE_SIZE),
-        VMSTATE_INT32_ARRAY_V(irq_state, PCIDevice, PCI_NUM_PINS, 2),
+        VMSTATE_BUFFER_UNSAFE_INFO(irq_state, PCIDevice, 2,
+				   vmstate_info_pci_irq_state,
+				   PCI_NUM_PINS * sizeof(int32_t)),
         VMSTATE_END_OF_LIST()
     }
 };
@@ -311,12 +396,23 @@ static inline const VMStateDescription *pci_get_vmstate(PCIDevice *s)
 
 void pci_device_save(PCIDevice *s, QEMUFile *f)
 {
+    /* Clear interrupt status bit: it is implicit
+     * in irq_state which we are saving.
+     * This makes us compatible with old devices
+     * which never set or clear this bit. */
+    s->config[PCI_STATUS] &= ~PCI_STATUS_INTERRUPT;
     vmstate_save_state(f, pci_get_vmstate(s), s);
+    /* Restore the interrupt status bit. */
+    pci_update_irq_status(s);
 }
 
 int pci_device_load(PCIDevice *s, QEMUFile *f)
 {
-    return vmstate_load_state(f, pci_get_vmstate(s), s, s->version_id);
+    int ret;
+    ret = vmstate_load_state(f, pci_get_vmstate(s), s, s->version_id);
+    /* Restore the interrupt status bit. */
+    pci_update_irq_status(s);
+    return ret;
 }
 
 static int pci_set_default_subsystem_id(PCIDevice *pci_dev)
@@ -330,6 +426,7 @@ static int pci_set_default_subsystem_id(PCIDevice *pci_dev)
 }
 
 /*
+ * Parse pci address in qemu command
  * Parse [[<domain>:]<bus>:]<slot>, return -1 on error
  */
 static int pci_parse_devaddr(const char *addr, int *domp, int *busp, unsigned *slotp)
@@ -378,6 +475,55 @@ static int pci_parse_devaddr(const char *addr, int *domp, int *busp, unsigned *s
     return 0;
 }
 
+/*
+ * Parse device bdf in device assignment command:
+ *
+ * -pcidevice host=bus:dev.func
+ *
+ * Parse <bus>:<slot>.<func> return -1 on error
+ */
+int pci_parse_host_devaddr(const char *addr, int *busp,
+                           int *slotp, int *funcp)
+{
+    const char *p;
+    char *e;
+    int val;
+    int bus = 0, slot = 0, func = 0;
+
+    p = addr;
+    val = strtoul(p, &e, 16);
+    if (e == p)
+	return -1;
+    if (*e == ':') {
+	bus = val;
+	p = e + 1;
+	val = strtoul(p, &e, 16);
+	if (e == p)
+	    return -1;
+	if (*e == '.') {
+	    slot = val;
+	    p = e + 1;
+	    val = strtoul(p, &e, 16);
+	    if (e == p)
+		return -1;
+	    func = val;
+	} else
+	    return -1;
+    } else
+	return -1;
+
+    if (bus > 0xff || slot > 0x1f || func > 0x7)
+	return -1;
+
+    if (*e)
+	return -1;
+
+    *busp = bus;
+    *slotp = slot;
+    *funcp = func;
+    return 0;
+}
+
 int pci_read_devaddr(Monitor *mon, const char *addr, int *domp, int *busp,
                      unsigned *slotp)
 {
@@ -490,16 +636,18 @@ static PCIDevice *do_pci_register_device(PCIDevice *pci_dev, PCIBus *bus,
             if (!bus->devices[devfn])
                 goto found;
         }
-        hw_error("PCI: no devfn available for %s, all in use\n", name);
+        qemu_error("PCI: no devfn available for %s, all in use\n", name);
+        return NULL;
     found: ;
     } else if (bus->devices[devfn]) {
-        hw_error("PCI: devfn %d not available for %s, in use by %s\n", devfn,
+        qemu_error("PCI: devfn %d not available for %s, in use by %s\n", devfn,
                  name, bus->devices[devfn]->name);
+        return NULL;
     }
     pci_dev->bus = bus;
     pci_dev->devfn = devfn;
     pstrcpy(pci_dev->name, sizeof(pci_dev->name), name);
-    memset(pci_dev->irq_state, 0, sizeof(pci_dev->irq_state));
+    pci_dev->irq_state = 0;
     pci_config_alloc(pci_dev);
 
     header_type &= ~PCI_HEADER_TYPE_MULTI_FUNCTION;
@@ -535,6 +683,9 @@ PCIDevice *pci_register_device(PCIBus *bus, const char *name,
     pci_dev = do_pci_register_device(pci_dev, bus, name, devfn,
                                      config_read, config_write,
                                      PCI_HEADER_TYPE_NORMAL);
+    if (pci_dev == NULL) {
+        hw_error("PCI: can't register device\n");
+    }
     return pci_dev;
 }
 static target_phys_addr_t pci_to_cpu_addr(target_phys_addr_t addr)
@@ -849,25 +1000,80 @@ static void pci_update_mappings(PCIDevice *d)
     }
 }
 
-uint32_t pci_default_read_config(PCIDevice *d,
+static uint32_t pci_read_config(PCIDevice *d,
-                                 uint32_t address, int len)
+                                uint32_t address, int len)
 {
     uint32_t val = 0;
-    assert(len == 1 || len == 2 || len == 4);
+
     len = MIN(len, pci_config_size(d) - address);
     memcpy(&val, d->config + address, len);
     return le32_to_cpu(val);
 }
 
+uint32_t pci_default_read_config(PCIDevice *d,
+                                 uint32_t address, int len)
+{
+    assert(len == 1 || len == 2 || len == 4);
+
+    if (pci_access_cap_config(d, address, len)) {
+        return d->cap.config_read(d, address, len);
+    }
+
+    return pci_read_config(d, address, len);
+}
+
+static void pci_write_config(PCIDevice *pci_dev,
+                             uint32_t address, uint32_t val, int len)
+{
+    int i;
+    for (i = 0; i < len; i++) {
+        pci_dev->config[address + i] = val & 0xff;
+        val >>= 8;
+    }
+}
+
+int pci_access_cap_config(PCIDevice *pci_dev, uint32_t address, int len)
+{
+    if (pci_dev->cap.supported && address >= pci_dev->cap.start &&
+            (address + len) < pci_dev->cap.start + pci_dev->cap.length)
+        return 1;
+    return 0;
+}
+
+uint32_t pci_default_cap_read_config(PCIDevice *pci_dev,
+                                     uint32_t address, int len)
+{
+    return pci_read_config(pci_dev, address, len);
+}
+
+void pci_default_cap_write_config(PCIDevice *pci_dev,
+                                  uint32_t address, uint32_t val, int len)
+{
+    pci_write_config(pci_dev, address, val, len);
+}
+
 void pci_default_write_config(PCIDevice *d, uint32_t addr, uint32_t val, int l)
 {
     int i;
     uint32_t config_size = pci_config_size(d);
 
+    if (pci_access_cap_config(d, addr, l)) {
+        d->cap.config_write(d, addr, val, l);
+        return;
+    }
+
     for (i = 0; i < l && addr + i < config_size; val >>= 8, ++i) {
         uint8_t wmask = d->wmask[addr + i];
         d->config[addr + i] = (d->config[addr + i] & ~wmask) | (val & wmask);
     }
+
+#ifdef CONFIG_KVM_DEVICE_ASSIGNMENT
+    if (kvm_enabled() && kvm_irqchip_in_kernel() &&
+        addr >= PIIX_CONFIG_IRQ_ROUTE &&
+	addr < PIIX_CONFIG_IRQ_ROUTE + 4)
+        assigned_dev_update_irqs();
+#endif /* CONFIG_KVM_DEVICE_ASSIGNMENT */
+
     if (ranges_overlap(addr, l, PCI_BASE_ADDRESS_0, 24) ||
         ranges_overlap(addr, l, PCI_ROM_ADDRESS, 4) ||
         ranges_overlap(addr, l, PCI_ROM_ADDRESS1, 4) ||
@@ -882,23 +1088,24 @@ void pci_default_write_config(PCIDevice *d, uint32_t addr, uint32_t val, int l)
 static void pci_set_irq(void *opaque, int irq_num, int level)
 {
     PCIDevice *pci_dev = opaque;
-    PCIBus *bus;
     int change;
 
-    change = level - pci_dev->irq_state[irq_num];
+    change = level - pci_irq_state(pci_dev, irq_num);
     if (!change)
         return;
 
-    pci_dev->irq_state[irq_num] = level;
+#if defined(TARGET_IA64)
-    for (;;) {
+    ioapic_set_irq(pci_dev, irq_num, level);
-        bus = pci_dev->bus;
+#endif
-        irq_num = bus->map_irq(pci_dev, irq_num);
+
-        if (bus->set_irq)
+    pci_set_irq_state(pci_dev, irq_num, level);
-            break;
+    pci_update_irq_status(pci_dev);
-        pci_dev = bus->parent_dev;
+    pci_change_irq_level(pci_dev, irq_num, change);
-    }
+}
-    bus->irq_count[irq_num] += change;
+
-    bus->set_irq(bus->irq_opaque, irq_num, bus->irq_count[irq_num] != 0);
+int pci_map_irq(PCIDevice *pci_dev, int pin)
+{
+    return pci_dev->bus->map_irq(pci_dev, pin);
 }
 
 /***********************************************************/
@@ -1270,9 +1477,17 @@ static int pci_qdev_init(DeviceState *qdev, DeviceInfo *base)
     pci_dev = do_pci_register_device(pci_dev, bus, base->name, devfn,
                                      info->config_read, info->config_write,
                                      info->header_type);
+    if (pci_dev == NULL)
+        return -1;
     rc = info->init(pci_dev);
     if (rc != 0)
         return rc;
+
+    /* rom loading */
+    if (pci_dev->romfile == NULL && info->romfile != NULL)
+        pci_dev->romfile = qemu_strdup(info->romfile);
+    pci_add_option_rom(pci_dev);
+
     if (qdev->hotplugged)
         bus->hotplug(pci_dev, 1);
     return 0;
@@ -1319,6 +1534,37 @@ PCIDevice *pci_create_simple(PCIBus *bus, int devfn, const char *name)
     return dev;
 }
 
+int pci_enable_capability_support(PCIDevice *pci_dev,
+                                  uint32_t config_start,
+                                  PCICapConfigReadFunc *config_read,
+                                  PCICapConfigWriteFunc *config_write,
+                                  PCICapConfigInitFunc *config_init)
+{
+    if (!pci_dev)
+        return -ENODEV;
+
+    pci_dev->config[0x06] |= 0x10; // status = capabilities
+
+    if (config_start == 0)
+	pci_dev->cap.start = PCI_CAPABILITY_CONFIG_DEFAULT_START_ADDR;
+    else if (config_start >= 0x40 && config_start < 0xff)
+        pci_dev->cap.start = config_start;
+    else
+        return -EINVAL;
+
+    if (config_read)
+        pci_dev->cap.config_read = config_read;
+    else
+        pci_dev->cap.config_read = pci_default_cap_read_config;
+    if (config_write)
+        pci_dev->cap.config_write = config_write;
+    else
+        pci_dev->cap.config_write = pci_default_cap_write_config;
+    pci_dev->cap.supported = 1;
+    pci_dev->config[PCI_CAPABILITY_LIST] = pci_dev->cap.start;
+    return config_init(pci_dev);
+}
+
 static int pci_find_space(PCIDevice *pdev, uint8_t size)
 {
     int config_size = pci_config_size(pdev);
@@ -1350,6 +1596,64 @@ static uint8_t pci_find_capability_list(PCIDevice *pdev, uint8_t cap_id,
     return next;
 }
 
+static void pci_map_option_rom(PCIDevice *pdev, int region_num, pcibus_t addr, pcibus_t size, int type)
+{
+    cpu_register_physical_memory(addr, size, pdev->rom_offset);
+}
+
+/* Add an option rom for the device */
+static int pci_add_option_rom(PCIDevice *pdev)
+{
+    int size;
+    char *path;
+    void *ptr;
+
+    if (!pdev->romfile)
+        return 0;
+    if (strlen(pdev->romfile) == 0)
+        return 0;
+
+    if (!pdev->rom_bar) {
+        /*
+         * Load rom via fw_cfg instead of creating a rom bar,
+         * for 0.11 compatibility.
+         */
+        int class = pci_get_word(pdev->config + PCI_CLASS_DEVICE);
+        if (class == 0x0300) {
+            rom_add_vga(pdev->romfile);
+        } else {
+            rom_add_option(pdev->romfile);
+        }
+        return 0;
+    }
+
+    path = qemu_find_file(QEMU_FILE_TYPE_BIOS, pdev->romfile);
+    if (path == NULL) {
+        path = qemu_strdup(pdev->romfile);
+    }
+
+    size = get_image_size(path);
+    if (size < 0) {
+        qemu_error("%s: failed to find romfile \"%s\"\n", __FUNCTION__,
+                   pdev->romfile);
+        return -1;
+    }
+    if (size & (size - 1)) {
+        size = 1 << qemu_fls(size);
+    }
+
+    pdev->rom_offset = qemu_ram_alloc(size);
+
+    ptr = qemu_get_ram_ptr(pdev->rom_offset);
+    load_image(path, ptr);
+    qemu_free(path);
+
+    pci_register_bar(pdev, PCI_ROM_SLOT, size,
+                     0, pci_map_option_rom);
+
+    return 0;
+}
+
 /* Reserve space and add capability to the linked list in pci config space */
 int pci_add_capability(PCIDevice *pdev, uint8_t cap_id, uint8_t size)
 {

hw/pci.h

@@ -5,11 +5,16 @@
 
 #include "qdev.h"
 
+struct kvm_irq_routing_entry;
+
 /* PCI includes legacy ISA access.  */
 #include "isa.h"
 
-/* PCI bus */
+/* imported from <linux/pci.h> */
+#define PCI_SLOT(devfn)         (((devfn) >> 3) & 0x1f)
+#define PCI_FUNC(devfn)         ((devfn) & 0x07)
 
+/* PCI bus */
 extern target_phys_addr_t pci_mem_base;
 
 #define PCI_DEVFN(slot, func)   ((((slot) & 0x1f) << 3) | ((func) & 0x07))
@@ -82,6 +87,12 @@ typedef void PCIMapIORegionFunc(PCIDevice *pci_dev, int region_num,
                                 pcibus_t addr, pcibus_t size, int type);
 typedef int PCIUnregisterFunc(PCIDevice *pci_dev);
 
+typedef void PCICapConfigWriteFunc(PCIDevice *pci_dev,
+                                   uint32_t address, uint32_t val, int len);
+typedef uint32_t PCICapConfigReadFunc(PCIDevice *pci_dev,
+                                      uint32_t address, int len);
+typedef int PCICapConfigInitFunc(PCIDevice *pci_dev);
+
 typedef struct PCIIORegion {
     pcibus_t addr; /* current PCI mapping address. -1 means not mapped */
 #define PCI_BAR_UNMAPPED (~(pcibus_t)0)
@@ -102,6 +113,7 @@ typedef struct PCIIORegion {
 #define  PCI_COMMAND_MEMORY	0x2	/* Enable response in Memory space */
 #define  PCI_COMMAND_MASTER	0x4	/* Enable bus master */
 #define PCI_STATUS              0x06    /* 16 bits */
+#define  PCI_STATUS_INTERRUPT   0x08
 #define PCI_REVISION_ID         0x08    /* 8 bits  */
 #define PCI_CLASS_PROG		0x09	/* Reg. Level Programming Interface */
 #define PCI_CLASS_DEVICE        0x0a    /* Device class */
@@ -159,10 +171,19 @@ typedef struct PCIIORegion {
 /* Bits in the PCI Status Register (PCI 2.3 spec) */
 #define PCI_STATUS_RESERVED1	0x007
 #define PCI_STATUS_INT_STATUS	0x008
+#ifndef PCI_STATUS_CAP_LIST
 #define PCI_STATUS_CAP_LIST	0x010
+#endif
+#ifndef PCI_STATUS_66MHZ
 #define PCI_STATUS_66MHZ	0x020
+#endif
+
 #define PCI_STATUS_RESERVED2	0x040
+
+#ifndef PCI_STATUS_FAST_BACK
 #define PCI_STATUS_FAST_BACK	0x080
+#endif
+
 #define PCI_STATUS_DEVSEL	0x600
 
 #define PCI_STATUS_RESERVED_MASK_LO (PCI_STATUS_RESERVED1 | \
@@ -191,6 +212,11 @@ enum {
     QEMU_PCI_CAP_EXPRESS = 0x2,
 };
 
+#define PCI_CAPABILITY_CONFIG_MAX_LENGTH 0x60
+#define PCI_CAPABILITY_CONFIG_DEFAULT_START_ADDR 0x40
+#define PCI_CAPABILITY_CONFIG_MSI_LENGTH 0x10
+#define PCI_CAPABILITY_CONFIG_MSIX_LENGTH 0x10
+
 struct PCIDevice {
     DeviceState qdev;
     /* PCI config space */
@@ -220,7 +246,7 @@ struct PCIDevice {
     qemu_irq *irq;
 
     /* Current IRQ levels.  Used internally by the generic PCI code.  */
-    int irq_state[PCI_NUM_PINS];
+    uint8_t irq_state;
 
     /* Capability bits */
     uint32_t cap_present;
@@ -241,6 +267,28 @@ struct PCIDevice {
     uint32_t msix_bar_size;
     /* Version id needed for VMState */
     int32_t version_id;
+
+    /* Location of option rom */
+    char *romfile;
+    ram_addr_t rom_offset;
+    uint32_t rom_bar;
+
+    /* How much space does an MSIX table need. */
+    /* The spec requires giving the table structure
+     * a 4K aligned region all by itself. Align it to
+     * target pages so that drivers can do passthrough
+     * on the rest of the region. */
+    target_phys_addr_t msix_page_size;
+
+    struct kvm_irq_routing_entry *msix_irq_entries;
+
+    /* Device capability configuration space */
+    struct {
+        int supported;
+        unsigned int start, length;
+        PCICapConfigReadFunc *config_read;
+        PCICapConfigWriteFunc *config_write;
+    } cap;
 };
 
 PCIDevice *pci_register_device(PCIBus *bus, const char *name,
@@ -252,6 +300,14 @@ void pci_register_bar(PCIDevice *pci_dev, int region_num,
                             pcibus_t size, int type,
                             PCIMapIORegionFunc *map_func);
 
+int pci_enable_capability_support(PCIDevice *pci_dev,
+                                  uint32_t config_start,
+                                  PCICapConfigReadFunc *config_read,
+                                  PCICapConfigWriteFunc *config_write,
+                                  PCICapConfigInitFunc *config_init);
+
+int pci_map_irq(PCIDevice *pci_dev, int pin);
+
 int pci_add_capability(PCIDevice *pci_dev, uint8_t cap_id, uint8_t cap_size);
 
 void pci_del_capability(PCIDevice *pci_dev, uint8_t cap_id, uint8_t cap_size);
@@ -260,13 +316,17 @@ void pci_reserve_capability(PCIDevice *pci_dev, uint8_t offset, uint8_t size);
 
 uint8_t pci_find_capability(PCIDevice *pci_dev, uint8_t cap_id);
 
-
 uint32_t pci_default_read_config(PCIDevice *d,
                                  uint32_t address, int len);
 void pci_default_write_config(PCIDevice *d,
                               uint32_t address, uint32_t val, int len);
 void pci_device_save(PCIDevice *s, QEMUFile *f);
 int pci_device_load(PCIDevice *s, QEMUFile *f);
+uint32_t pci_default_cap_read_config(PCIDevice *pci_dev,
+                                     uint32_t address, int len);
+void pci_default_cap_write_config(PCIDevice *pci_dev,
+                                  uint32_t address, uint32_t val, int len);
+int pci_access_cap_config(PCIDevice *pci_dev, uint32_t address, int len);
 
 typedef void (*pci_set_irq_fn)(void *opaque, int irq_num, int level);
 typedef int (*pci_map_irq_fn)(PCIDevice *pci_dev, int irq_num);
@@ -295,6 +355,9 @@ PCIBus *pci_get_bus_devfn(int *devfnp, const char *devaddr);
 int pci_read_devaddr(Monitor *mon, const char *addr, int *domp, int *busp,
                      unsigned *slotp);
 
+int pci_parse_host_devaddr(const char *addr, int *busp,
+                           int *slotp, int *funcp);
+
 void pci_info(Monitor *mon);
 PCIBus *pci_bridge_init(PCIBus *bus, int devfn, uint16_t vid, uint16_t did,
                         pci_map_irq_fn map_irq, const char *name);
@@ -379,6 +442,9 @@ typedef struct {
 
     /* pcie stuff */
     int is_express;   /* is this device pci express? */
+
+    /* rom bar */
+    const char *romfile;
 } PCIDeviceInfo;
 
 void pci_qdev_register(PCIDeviceInfo *info);

hw/pcspk.c

@@ -27,6 +27,8 @@
 #include "isa.h"
 #include "audio/audio.h"
 #include "qemu-timer.h"
+#include "i8254.h"
+#include "qemu-kvm.h"
 
 #define PCSPK_BUF_LEN 1792
 #define PCSPK_SAMPLE_RATE 32000
@@ -48,6 +50,43 @@ typedef struct {
 static const char *s_spk = "pcspk";
 static PCSpkState pcspk_state;
 
+#ifdef CONFIG_KVM_PIT
+static void kvm_get_pit_ch2(PITState *pit,
+                            struct kvm_pit_state *inkernel_state)
+{
+    struct kvm_pit_state pit_state;
+
+    if (kvm_enabled() && qemu_kvm_pit_in_kernel()) {
+        kvm_get_pit(kvm_context, &pit_state);
+        pit->channels[2].mode = pit_state.channels[2].mode;
+        pit->channels[2].count = pit_state.channels[2].count;
+        pit->channels[2].count_load_time = pit_state.channels[2].count_load_time;
+        pit->channels[2].gate = pit_state.channels[2].gate;
+        if (inkernel_state) {
+            memcpy(inkernel_state, &pit_state, sizeof(*inkernel_state));
+        }
+    }
+}
+
+static void kvm_set_pit_ch2(PITState *pit,
+                            struct kvm_pit_state *inkernel_state)
+{
+    if (kvm_enabled() && qemu_kvm_pit_in_kernel()) {
+        inkernel_state->channels[2].mode = pit->channels[2].mode;
+        inkernel_state->channels[2].count = pit->channels[2].count;
+        inkernel_state->channels[2].count_load_time =
+            pit->channels[2].count_load_time;
+        inkernel_state->channels[2].gate = pit->channels[2].gate;
+        kvm_set_pit(kvm_context, inkernel_state);
+    }
+}
+#else
+static inline void kvm_get_pit_ch2(PITState *pit,
+                                   struct kvm_pit_state *inkernel_state) { }
+static inline void kvm_set_pit_ch2(PITState *pit,
+                                   struct kvm_pit_state *inkernel_state) { }
+#endif
+
 static inline void generate_samples(PCSpkState *s)
 {
     unsigned int i;
@@ -72,6 +111,8 @@ static void pcspk_callback(void *opaque, int free)
     PCSpkState *s = opaque;
     unsigned int n;
 
+    kvm_get_pit_ch2(s->pit, NULL);
+
     if (pit_get_mode(s->pit, 2) != 3)
         return;
 
@@ -117,6 +158,8 @@ static uint32_t pcspk_ioport_read(void *opaque, uint32_t addr)
     PCSpkState *s = opaque;
     int out;
 
+    kvm_get_pit_ch2(s->pit, NULL);
+
     s->dummy_refresh_clock ^= (1 << 4);
     out = pit_get_out(s->pit, 2, qemu_get_clock(vm_clock)) << 5;
 
@@ -125,9 +168,12 @@ static uint32_t pcspk_ioport_read(void *opaque, uint32_t addr)
 
 static void pcspk_ioport_write(void *opaque, uint32_t addr, uint32_t val)
 {
+    struct kvm_pit_state inkernel_state;
     PCSpkState *s = opaque;
     const int gate = val & 1;
 
+    kvm_get_pit_ch2(s->pit, &inkernel_state);
+
     s->data_on = (val >> 1) & 1;
     pit_set_gate(s->pit, 2, gate);
     if (s->voice) {
@@ -135,6 +181,8 @@ static void pcspk_ioport_write(void *opaque, uint32_t addr, uint32_t val)
             s->play_pos = 0;
         AUD_set_active_out(s->voice, gate & s->data_on);
     }
+
+    kvm_set_pit_ch2(s->pit, &inkernel_state);
 }
 
 void pcspk_init(PITState *pit)

hw/piix_pci.c

@@ -29,6 +29,8 @@
 #include "isa.h"
 #include "sysbus.h"
 
+#include "qemu-kvm.h"
+
 typedef PCIHostState I440FXState;
 
 typedef struct PIIX3State {
@@ -88,6 +90,10 @@ static void i440fx_update_memory_mappings(PCII440FXState *d)
     int i, r;
     uint32_t smram, addr;
 
+    if (kvm_enabled()) {
+        /* FIXME: Support remappings and protection changes. */
+        return;
+    }
     update_pam(d, 0xf0000, 0x100000, (d->dev.config[0x59] >> 4) & 3);
     for(i = 0; i < 12; i++) {
         r = (d->dev.config[(i >> 1) + 0x5a] >> ((i & 1) * 4)) & 3;
@@ -201,6 +207,8 @@ static int i440fx_initfn(PCIDevice *dev)
     return 0;
 }
 
+static PIIX3State *piix3_dev;
+
 PCIBus *i440fx_init(PCII440FXState **pi440fx_state, int *piix3_devfn, qemu_irq *pic)
 {
     DeviceState *dev;
@@ -226,6 +234,8 @@ PCIBus *i440fx_init(PCII440FXState **pi440fx_state, int *piix3_devfn, qemu_irq *
 
     *piix3_devfn = piix3->dev.devfn;
 
+    piix3_dev = piix3;
+
     return b;
 }
 
@@ -253,6 +263,13 @@ static void piix3_set_irq(void *opaque, int irq_num, int level)
     }
 }
 
+int piix_get_irq(int pin)
+{
+    if (piix3_dev)
+        return piix3_dev->dev.config[0x60+pin];
+    return 0;
+}
+
 static void piix3_reset(void *opaque)
 {
     PIIX3State *d = opaque;

hw/ppc440.c

@@ -20,6 +20,7 @@
 #include "ppc405.h"
 #include "sysemu.h"
 #include "kvm.h"
+#include "qemu-kvm.h"
 
 #define PPC440EP_PCI_CONFIG     0xeec00000
 #define PPC440EP_PCI_INTACK     0xeed00000

hw/ppc440_bamboo.c

@@ -24,6 +24,7 @@
 #include "device_tree.h"
 #include "loader.h"
 #include "elf.h"
+#include "qemu-kvm.h"
 
 #define BINARY_DEVICE_TREE_FILE "bamboo.dtb"
 

hw/ppce500_mpc8544ds.c

@@ -31,6 +31,7 @@
 #include "ppce500.h"
 #include "loader.h"
 #include "elf.h"
+#include "qemu-kvm.h"
 
 #define BINARY_DEVICE_TREE_FILE    "mpc8544ds.dtb"
 #define UIMAGE_LOAD_BASE           0

hw/qdev-properties.c

@@ -500,7 +500,12 @@ int qdev_prop_parse(DeviceState *dev, const char *name, const char *value)
                 dev->info->name, name);
         return -1;
     }
-    return prop->info->parse(dev, prop, value);
+    if (prop->info->parse(dev, prop, value) != 0) {
+        fprintf(stderr, "property \"%s.%s\": failed to parse \"%s\"\n",
+                dev->info->name, name, value);
+        return -1;
+    }
+    return 0;
 }
 
 void qdev_prop_set(DeviceState *dev, const char *name, void *src, enum PropertyType type)
@@ -593,26 +598,33 @@ void qdev_prop_set_defaults(DeviceState *dev, Property *props)
     }
 }
 
-static CompatProperty *compat_props;
+static QTAILQ_HEAD(, GlobalProperty) global_props = QTAILQ_HEAD_INITIALIZER(global_props);
 
-void qdev_prop_register_compat(CompatProperty *props)
+void qdev_prop_register_global(GlobalProperty *prop)
 {
-    compat_props = props;
+    QTAILQ_INSERT_TAIL(&global_props, prop, next);
 }
 
-void qdev_prop_set_compat(DeviceState *dev)
+void qdev_prop_register_global_list(GlobalProperty *props)
 {
-    CompatProperty *prop;
+    int i;
 
-    if (!compat_props) {
+    for (i = 0; props[i].driver != NULL; i++) {
-        return;
+        qdev_prop_register_global(props+i);
     }
-    for (prop = compat_props; prop->driver != NULL; prop++) {
+}
-        if (strcmp(dev->info->name, prop->driver) != 0) {
+
+void qdev_prop_set_globals(DeviceState *dev)
+{
+    GlobalProperty *prop;
+
+    QTAILQ_FOREACH(prop, &global_props, next) {
+        if (strcmp(dev->info->name, prop->driver) != 0 &&
+            strcmp(dev->info->bus_info->name, prop->driver) != 0) {
             continue;
         }
         if (qdev_prop_parse(dev, prop->property, prop->value) != 0) {
-            abort();
+            exit(1);
         }
     }
 }

hw/qdev.c

@@ -103,7 +103,7 @@ DeviceState *qdev_create(BusState *bus, const char *name)
     dev->parent_bus = bus;
     qdev_prop_set_defaults(dev, dev->info->props);
     qdev_prop_set_defaults(dev, dev->parent_bus->info->props);
-    qdev_prop_set_compat(dev);
+    qdev_prop_set_globals(dev);
     QLIST_INSERT_HEAD(&bus->children, dev, sibling);
     if (qdev_hotplug) {
         assert(bus->allow_hotplug);

hw/qdev.h

@@ -92,11 +92,12 @@ struct PropertyInfo {
     int (*print)(DeviceState *dev, Property *prop, char *dest, size_t len);
 };
 
-struct CompatProperty {
+typedef struct GlobalProperty {
     const char *driver;
     const char *property;
     const char *value;
-};
+    QTAILQ_ENTRY(GlobalProperty) next;
+} GlobalProperty;
 
 /*** Board API.  This should go away once we have a machine config file.  ***/
 
@@ -256,8 +257,9 @@ void qdev_prop_set_macaddr(DeviceState *dev, const char *name, uint8_t *value);
 void qdev_prop_set_ptr(DeviceState *dev, const char *name, void *value);
 void qdev_prop_set_defaults(DeviceState *dev, Property *props);
 
-void qdev_prop_register_compat(CompatProperty *props);
+void qdev_prop_register_global(GlobalProperty *prop);
-void qdev_prop_set_compat(DeviceState *dev);
+void qdev_prop_register_global_list(GlobalProperty *props);
+void qdev_prop_set_globals(DeviceState *dev);
 
 /* This is a nasty hack to allow passing a NULL bus to qdev_create.  */
 extern struct BusInfo system_bus_info;

hw/rtl8139.c

@@ -3353,14 +3353,6 @@ static int pci_rtl8139_init(PCIDevice *dev)
     qemu_mod_timer(s->timer,
         rtl8139_get_next_tctr_time(s,qemu_get_clock(vm_clock)));
 #endif /* RTL8139_ONBOARD_TIMER */
-
-    if (!dev->qdev.hotplugged) {
-        static int loaded = 0;
-        if (!loaded) {
-            rom_add_option("pxe-rtl8139.bin");
-            loaded = 1;
-        }
-    }
     return 0;
 }
 
@@ -3371,6 +3363,7 @@ static PCIDeviceInfo rtl8139_info = {
     .qdev.vmsd  = &vmstate_rtl8139,
     .init       = pci_rtl8139_init,
     .exit       = pci_rtl8139_uninit,
+    .romfile    = "pxe-rtl8139.bin",
     .qdev.props = (Property[]) {
         DEFINE_NIC_PROPERTIES(RTL8139State, conf),
         DEFINE_PROP_END_OF_LIST(),

hw/s390-virtio-bus.c

@@ -307,7 +307,7 @@ static void virtio_s390_notify(void *opaque, uint16_t vector)
     uint64_t token = s390_virtio_device_vq_token(dev, vector);
 
     /* XXX kvm dependency! */
-    kvm_s390_virtio_irq(s390_cpu_addr2state(0), 1, token);
+    kvm_s390_virtio_irq(s390_cpu_addr2state(0), 0, token);
 }
 
 /**************** S390 Virtio Bus Device Descriptions *******************/

hw/s390-virtio.c

@@ -142,6 +142,13 @@ static void s390_init(ram_addr_t ram_size,
     ram_addr_t initrd_size = 0;
     int i;
 
+    /* XXX we only work on KVM for now */
+
+    if (!kvm_enabled()) {
+        fprintf(stderr, "The S390 target only works with KVM enabled\n");
+        exit(1);
+    }
+
     /* get a BUS */
     s390_bus = s390_virtio_bus_init(&ram_size);
 
@@ -181,7 +188,7 @@ static void s390_init(ram_addr_t ram_size,
 
         cpu_synchronize_state(env);
         env->psw.addr = KERN_IMAGE_START;
-        env->psw.mask = 0x0000000180000000UL;
+        env->psw.mask = 0x0000000180000000ULL;
     }
 
     if (initrd_filename) {
@@ -201,7 +208,11 @@ static void s390_init(ram_addr_t ram_size,
     }
 
     /* Create VirtIO console */
-    qdev_init_nofail(qdev_create((BusState *)s390_bus, "virtio-console-s390"));
+    for(i = 0; i < MAX_VIRTIO_CONSOLES; i++) {
+        if (virtcon_hds[i]) {
+            qdev_init_nofail(qdev_create((BusState *)s390_bus, "virtio-console-s390"));
+        }
+    }
 
     /* Create VirtIO network adapters */
     for(i = 0; i < nb_nics; i++) {
@@ -209,7 +220,7 @@ static void s390_init(ram_addr_t ram_size,
         DeviceState *dev;
 
         if (!nd->model) {
-            nd->model = (char*)"virtio";
+            nd->model = qemu_strdup("virtio");
         }
 
         if (strcmp(nd->model, "virtio")) {
@@ -243,6 +254,10 @@ static QEMUMachine s390_machine = {
     .alias = "s390",
     .desc = "VirtIO based S390 machine",
     .init = s390_init,
+    .no_serial = 1,
+    .no_parallel = 1,
+    .use_virtcon = 1,
+    .no_vga = 1,
     .max_cpus = 255,
     .is_default = 1,
 };