Anthony Liguori [Fri, 26 Jul 2013 18:04:07 +0000 (13:04 -0500)]
Merge remote-tracking branch 'jliu/or32' into staging
# By Jia Liu
# Via Jia Liu
* jliu/or32:
target-openrisc: Free typename in openrisc_cpu_class_by_name
hw/openrisc: Use stderr output instead of qemu_log
hw/openrisc: Indent typo
Message-id: 1374576458-22808-1-git-send-email-proljc@gmail.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Anthony Liguori [Thu, 25 Jul 2013 20:58:29 +0000 (15:58 -0500)]
Merge remote-tracking branch 'mdroth/qga-pull-2013-7-25' into staging
# By Laszlo Ersek
# Via Michael Roth
* mdroth/qga-pull-2013-7-25:
qga: escape cmdline args when registering win32 service (CVE-2013-2231)
ga_install_service(): nest error paths more idiomatically
qga/service-win32.c: diagnostic output should go to stderr
Anthony Liguori [Thu, 25 Jul 2013 20:56:06 +0000 (15:56 -0500)]
Merge remote-tracking branch 'riku/linux-user-for-upstream' into staging
* riku/linux-user-for-upstream: (21 commits)
linux-user: Handle compressed ISA encodings when processing MIPS exceptions
linux-user: Unlock mmap_lock when resuming guest from page_unprotect
linux-user: Reset copied CPUs in cpu_copy() always
linux-user: Fix epoll on ARM hosts
linux-user: fix segmentation fault passing with h2g(x) != x
linux-user: Fix pipe syscall return for SPARC
linux-user: Fix target_stat and target_stat64 for OpenRISC
linux-user: Avoid conditional cpu_reset()
configure: Make NPTL non-optional
linux-user: Enable NPTL for x86-64
linux-user: Add i386 TLS setter
linux-user: Clean up handling of clone() argument order
linux-user: Add missing 'break' in i386 get_thread_area syscall
linux-user: Enable NPTL for m68k
linux-user: Enable NPTL for SPARC targets
linux-user: Enable NPTL for OpenRISC
linux-user: Move includes of target-specific headers to end of qemu.h
configure: Enable threading for unicore32-linux-user
configure: Enable threading on all ppc and mips linux-user targets
configure: Don't say target_nptl="no" if there is no linux-user target
...
Conflicts:
linux-user/main.c
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:38 +0000 (15:54 +0200)]
pc-testdev: add I/O port to test memory.c auto split/combine
The ports at 0xe8..0xeb have impl.min/max_access_size == 1, so
that memory accesses are split and combined by the memory core.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-29-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:37 +0000 (15:54 +0200)]
memory: check memory region endianness, not target's
When combining multiple accesses into a single value, we need to do so
in the device's desired endianness. The target endianness does not have
any influence.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-28-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:36 +0000 (15:54 +0200)]
memory: pass MemoryRegion to access_with_adjusted_size
The accessors all use a MemoryRegion opaque value. Avoid going
uselessly through void*.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-27-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:35 +0000 (15:54 +0200)]
memory: move functions around
Prepare for next patch, no semantic change.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-26-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:34 +0000 (15:54 +0200)]
qtest: add test for ISA I/O space endianness
This writes a register and reads its 1/2/4 byte parts. Masking
is done in the device model.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-25-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:33 +0000 (15:54 +0200)]
default-configs/ppc64: add all components of i82378 SuperIO chip used by prep
The device provides an ISA bus so that pseries can also run the
endianness test.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-24-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:32 +0000 (15:54 +0200)]
default-configs: add SuperIO to SH4
The device provides an ISA bus to run the endianness test on.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-23-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:31 +0000 (15:54 +0200)]
default-configs: add test device to all machines supporting ISA
This will let these machines run an endianness test for ISA
I/O port space.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-22-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:30 +0000 (15:54 +0200)]
sparc64: unbreak
... by making apb a subclass of TYPE_PCI_HOST_BRIDGE.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-21-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:29 +0000 (15:54 +0200)]
sh4: unbreak r2d
... by making sh_pci a subclass of TYPE_PCI_HOST_BRIDGE.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-20-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:28 +0000 (15:54 +0200)]
mips: degrade BIOS error to warning
No free MIPS BIOS is available, so it makes little sense to quit.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-19-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
So far the device was only used on little-endian machines.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-18-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:26 +0000 (15:54 +0200)]
pc-testdev: support 8 and 16-bit accesses to 0xe0
This will let us use the testdev to test endianness.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-17-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
The scenario where I/O ports are accessed with DEVICE_LITTLE_ENDIAN
endianness now works and will soon be unit tested. Since the PortioList
indirection assumes little endian, define portio_ops the same way.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-16-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:24 +0000 (15:54 +0200)]
isa_mmio: delete
It is not used anymore.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-15-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:23 +0000 (15:54 +0200)]
ebus: do not use isa_mmio
This is untested, because ebus does not have a libqos module.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-14-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:22 +0000 (15:54 +0200)]
sparc64: remove indirection for I/O port access
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-13-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:21 +0000 (15:54 +0200)]
mips_fulong2e: do not use isa_mmio
This board is little-endian, but still isa_mmio should die. :)
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-12-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:20 +0000 (15:54 +0200)]
mipssim: do not use isa_mmio
Untested, this board does not support PCI so it cannot run endianness-test.
It should fix endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-11-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:19 +0000 (15:54 +0200)]
ppc440_bamboo: do not use isa_mmio
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-10-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:18 +0000 (15:54 +0200)]
mips_malta: do not use isa_mmio
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-9-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:17 +0000 (15:54 +0200)]
mips_r4k: do not use isa_mmio
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-8-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:16 +0000 (15:54 +0200)]
mips_jazz: do not use isa_mmio
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-7-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:15 +0000 (15:54 +0200)]
prep: fix I/O port endianness
Do not swap endianness here, it will happen during cpu_{in,out}{b,w,l}.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-6-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:13 +0000 (15:54 +0200)]
ppc_newworld: do not use isa_mmio
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-4-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:12 +0000 (15:54 +0200)]
ppc_oldworld: do not use isa_mmio
This fixes endianness bugs in I/O port access.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-3-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Paolo Bonzini [Mon, 22 Jul 2013 13:54:11 +0000 (15:54 +0200)]
sh4: do not use isa_mmio
This fixes endianness bugs in I/O port access (for sh4eb).
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Anthony Liguori <aliguori@us.ibm.com>
Message-id: 1374501278-31549-2-git-send-email-pbonzini@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Alex Williamson (4):
seabios q35: Enable all PIRQn IRQs at startup
seabios q35: Add new PCI slot to irq routing function
seabios: Add a dummy PCI slot to irq mapping function
pciinit: Enable default VGA device
Asias He (2):
virtio-scsi: Set _DRIVER_OK flag before scsi target scanning
virtio-scsi: Pack struct virtio_scsi_{req_cmd,resp_cmd}
Avik Sil (1):
USB-EHCI: Fix null pointer assignment
Christian Gmeiner (5):
geodevga: fix errors in geode_fp_* functions
geodevga: move framebuffer setup
geodevga: move output setup to own function
geodevga: add debug to msr functions
geodevga: fix wrong define name
David Woodhouse (26):
Add macros for pushing and popping struct bregs
Clean up #if in pirtable.c. CONFIG_PIRTABLE can't be set if CONFIG_COREBOOT is
post: Export functions which will be used individually by CSM
Export callrom() for CSM to use
Export copy_smbios() from biostables.c
Import LegacyBios.h from OVMF
Complete and checksum EFI_COMPATIBILITY16_TABLE at build time
Add pic_save_mask() and pic_restore_mask() functions
Add CSM support
Add README.CSM
Add find_pmtimer() function
Enable PMTIMER for CSM build
Fix rom_reserve()/rom_confirm() for CSM oprom dispatch
Don't calibrate TSC if PMTIMER is already set up
Move find_pmtimer() to ACPI table setup where it logically belongs
Use find_pmtimer() after copying Xen ACPI tables
Use find_pmtimer() after copying coreboot ACPI tables
Unify return path for CSM to go via csm_return()
Make CONFIG_OPTIONROMS_DEPLOYED depend on CONFIG_QEMU
Implement !CONFIG_OPTIONROMS support for CSM
Implement !CONFIG_BOOT for CSM
Enable VGA output when settings bochs-specific mode
Disable CONFIG_THREAD_OPTIONROMS for CSM build
Fix return type of le64_to_cpu() and be64_to_cpu()
Rename find_pmtimer() to find_acpi_features()
Add acpi_reboot() reset method using RESET_REG
Kevin O'Connor (99):
pmm: Use 'struct segoff_s' in pmm header.
Minor: Update README - variable changes are now reset on soft-reboots.
Normalize POST initialization function name suffixes.
POST: Reorganize post init functions for better grouping and reusability.
Fix rebase error in commit 8a0a972f that broke LOWMEM variables.
Support calling a function other than maininit() from reloc_preinit().
Ensure exported symbols are visible in the final link
POST: Move QEMU specific ramsize and BIOS table setup to paravirt.c.
POST: Reorganize post entry and "preinit" functions.
POST: Move cpu caching and dma setup to platform_hardware_setup().
Undo incorrect assumptions about Xen in commit 6ca0460f.
Determine century during init and store in VARLOW mem during runtime.
No need to check both CONFIG_THREADS and CONFIG_THREAD_OPTIONROMS.
Add runningOnQEMU() and runningOnXen() for runtime platform detection.
Consistently use CONFIG_COREBOOT, CONFIG_QEMU, and runningOnXen().
Convert kvm_para_available() to runningOnKVM().
Minor - move definitions to paravirt.c from paravirt.h.
Only perform SMP setup on QEMU.
Start device_hardware_setup in mainint even with CONFIG_THREAD_OPTIONROMS.
The mathcp setup touches the PIC and thus move to the "setup" phase.
Update tools/acpi_extract.py to handle iasl 20130117 release.
Support skipping content when reading from QEMU fw_cfg romfile entries.
Convert fw_cfg ACPI entries into romfile entries.
Convert fw_cfg SMBIOS entries into romfile entries.
Convert basic integer fw_cfg entries into romfile entries.
Convert fw_cfg NUMA entries into a romfile entry.
Process fw_cfg e820 entries during the fw_cfg setup stage.
Integrate qemu_cfg_preinit() into qemu_romfile_init().
Group QEMU platform setup together and move to paravirt.c.
vgabios: Bochs/QEMU vgabios support should depend on CONFIG_QEMU.
Warn on unaligned PCI ROM structure in option roms.
Fix Makefile - don't reference "out/" directly, instead use "$(OUT)".
build: Don't require $(OUT) to be a sub-directory of the main directory.
Rename rom_get_top() to rom_get_max().
Report on f-segment UMB ram also.
Clarify build generated "zone low" values.
Verify CC is valid during build tests.
Disable handle_post() on CSM builds.
Remove unnecessary "export" declarations from assembler functions.
Minor assembler enhancements to __csm_return.
Introduce VARFSEG for variables that will reside in the f-segment.
Convert VAR16VISIBLE, VAR16EXPORT, and VAR32VISIBLE to VARFSEG.
Don't relocate "varlow" variable references at runtime.
Move malloc's ZoneFSeg and ZoneLow setup to malloc_init.
Calculate "RamSize" needed by 16bit interface dynamically.
Eliminate separate BiosTableSpace[] space for f-segment allocations.
Use CONFIG_ prefix for Kconfig variables; use BUILD_ for others.
Try to detect an unsuccessful hard-reboot to prevent soft-reboot loops.
Minor - fix confusing final_sec32low_start name in layoutrom.py.
Minor - introduce numeric defines for the IVT offset of hw irqs.
Separate out 16bit PCI-BIOS entry point from regular int 0x1a entry point.
Support using the "extra stack" for all 16bit irq entry points.
Minor - improve comments and grouping of handle_08().
floppy: Introduce 'struct floppy_pio_s' for floppy PIO ops.
floppy: Cleanup floppy irq wait handling.
floppy: Clean up Check Interrupt Status code.
floppy: Move recalibration and results parsing to floppy_cmd().
floppy: Improve floppy_pio() error checking.
floppy: Implement media format sensing.
floppy: Actually do controller reset in floppy_reset().
Minor - note that passing QEMU config via cmos is deprecated.
Cache boot-fail-wait to avoid romfile access after POST.
Rename src/ssdt-susp.dsl to src/ssdt-misc.dsl.
acpi: Eliminate BDAT parameter passing to DSDT code.
Add additional dependency checks to Makefile.
Don't use __FILE__ in virtio-ring.c.
shadow: Don't use PCIDevices list in make_bios_readonly().
smm: Don't use PCIDevices list in smm_setup().
Add VARVERIFY32INIT attribute for variables only available during "init".
Use VARVERIFY32INIT on global variables that point to "tmp" memory.
vgabios: Fix stdvga_perform_gray_scale_summing().
vgabios: Fix cirrus memory clear on mode switch.
Minor - add missing newline to floppy debug statement.
Fix bug in NUMA node setup - don't create SRAT if NUMA not present.
Update README - copy *.aml files for QEMU.
Add dependencies to vgafixup.py and buildversion.sh scripts.
Set ZF prior to keyboard read call in check_for_keystroke().
mptable: Don't describe pci-to-pci bridges.
mptable: Use same PCI irqs as ACPI code.
Cleanup QEMU_CFG_NUMA fw_cfg processing - split into two romfile entries.
Use container_of on romfile entries.
acpi: Move ACPI table definitions from acpi.c to acpi.h.
acpi: Remove dead code with descriptions of bit flags.
acpi: Use cpu_to_leXX() consistently.
Minor - explicitly close files in buildrom.py.
Minor - move "tracked memory alloc" code in pmm.c.
Introduce and convert pmm code to use standard list helpers.
Minor - relocate code in stacks.c to keep low-level thread code together.
Introduce helper function have_threads() in stacks.c.
Convert stacks.c to use standard list manipulation code.
Convert boot.c to use standard list manipulation code.
Convert pciinit.c to use standard list manipulation code.
Convert PCIDevices list to use standard list manipultion code.
Revert "Convert pciinit.c to use standard list manipulation code."
Fix error in hlist_for_each_entry_safe macro.
Convert pciinit.c to use standard list manipulation code.
make qemu_cfg_init depend on QEMU_HARDWARE instead of QEMU
Another fix for hlist_for_each_entry_safe.
Minor - remove debugging dprintf added to pciinit.c.
Laszlo Ersek (1):
Enable VGA output when setting Cirrus-specific mode
Michael S. Tsirkin (1):
acpi: make default DSDT optional
Stefan Weil [Fri, 5 Jul 2013 20:55:42 +0000 (22:55 +0200)]
po/Makefile: Fix *.mo generation for out-of-tree builds (regression)
Commit f84756554e32d97db3aa949db1dd58c7eea62375 added a wildcard search
for *.po files. This search found no files for out of tree builds, so
those builds no longer created and installed *.mo files.
Signed-off-by: Stefan Weil <sw@weilnetz.de> Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Anthony Liguori [Tue, 23 Jul 2013 15:57:23 +0000 (10:57 -0500)]
Merge remote-tracking branch 'quintela/migration.next' into staging
# By Michael R. Hines (8) and others
# Via Juan Quintela
* quintela/migration.next:
migration: add autoconvergence documentation
Fix real mode guest segments dpl value in savevm
Fix real mode guest migration
rdma: account for the time spent in MIG_STATE_SETUP through QMP
rdma: introduce MIG_STATE_NONE and change MIG_STATE_SETUP state transition
rdma: allow state transitions between other states besides ACTIVE
rdma: send pc.ram
rdma: core logic
rdma: introduce ram_handle_compressed()
rdma: bugfix: ram_control_save_page()
rdma: update documentation to reflect new unpin support
Message-id: 1374590725-14144-1-git-send-email-quintela@redhat.com Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
Anthony Liguori [Tue, 23 Jul 2013 15:57:04 +0000 (10:57 -0500)]
Merge remote-tracking branch 'afaerber/tags/qom-cpu-for-anthony' into staging
QOM CPUState refactorings
* Fix NULL pointer dereference in gdbstub
* Introduce vaddr type
* Introduce CPUClass::set_pc()
* Introduce CPUClass::synchronize_from_tb()
* Introduce CPUClass::get_phys_page_debug()
* Introduce CPUClass::memory_rw_debug()
* Move singlestep_enabled and gdb_regs fields out of CPU_COMMON
* Adopt CPUState in more APIs
* Propagate CPUState in gdbstub
# gpg: Signature made Mon 22 Jul 2013 07:50:17 PM CDT using RSA key ID 3E7E013F
# gpg: Can't check signature: public key not found
# By Andreas Färber (21) and others
# Via Andreas Färber
* afaerber/tags/qom-cpu-for-anthony: (24 commits)
linux-user: Use X86CPU property to retrieve CPUID family
gdbstub: Change gdb_register_coprocessor() argument to CPUState
cpu: Move gdb_regs field from CPU_COMMON to CPUState
gdbstub: Change GDBState::{c,g}_cpu and find_cpu() to CPUState
cpu: Introduce CPUClass::memory_rw_debug() for target_memory_rw_debug()
exec: Change cpu_memory_rw_debug() argument to CPUState
cpu: Turn cpu_get_phys_page_debug() into a CPUClass hook
gdbstub: Change gdb_{read,write}_register() argument to CPUState
gdbstub: Change gdb_handlesig() argument to CPUState
gdbstub: Change syscall callback argument to CPUState
kvm: Change kvm_{insert,remove}_breakpoint() argument to CPUState
cpu: Change cpu_single_step() argument to CPUState
gdbstub: Update gdb_handlesig() and gdb_signalled() Coding Style
cpu: Move singlestep_enabled field from CPU_COMMON to CPUState
target-alpha: Copy implver to DisasContext
target-alpha: Copy singlestep_enabled to DisasContext
cpu: Introduce CPUClass::synchronize_from_tb() for cpu_pc_from_tb()
target-unicore32: Implement CPUClass::set_pc()
target-moxie: Implement CPUClass::set_pc()
target-m68k: Implement CPUClass::set_pc()
...
Anthony Liguori [Tue, 23 Jul 2013 15:56:55 +0000 (10:56 -0500)]
Merge remote-tracking branch 'afaerber/tags/qom-devices-for-anthony' into staging
QOM device refactorings
* Avoid TYPE_* in VMStateDescription name
* Replace some DO_UPCAST()s and FROM_SYSBUS()s with QOM casts
* Limit legacy SCSI command line handling to non-hotplugged devices
* Replace some SysBusDeviceClass::init with DeviceClass::realize
# gpg: Signature made Mon 22 Jul 2013 06:31:42 PM CDT using RSA key ID 3E7E013F
# gpg: Can't check signature: public key not found
# By Hu Tao (26) and others
# Via Andreas Färber
* afaerber/tags/qom-devices-for-anthony: (55 commits)
isa-bus: Drop isabus_bridge_init() since it does nothing
ioapic: Use QOM realize for ioapic
kvmvapic: Use QOM realize
kvm/clock: Use QOM realize for kvmclock
hpet: Use QOM realize for hpet
scsi: Improve error propagation for scsi_bus_legacy_handle_cmdline()
megasas: Legacy command line handling fix
scsi/esp: Use QOM realize for scsi esp
fw_cfg: Use QOM realize for fw_cfg
ahci: Use QOM realize for ahci
pflash_cfi02: Use QOM realize for pflash_cfi02
pflash_cfi01: Use QOM realize for pflash_cfi01
fdc: Improve error propagation for QOM realize
fdc: Use QOM realize for fdc
kvm/clock: QOM'ify some more
hpet: QOM'ify some more
scsi/esp: QOM'ify some more
fwcfg: QOM'ify some more
ahci: QOM'ify some more
pflash-cfi02: QOM'ify some more
...
Alexander Graf [Sat, 6 Jul 2013 12:17:57 +0000 (14:17 +0200)]
linux-user: Unlock mmap_lock when resuming guest from page_unprotect
The page_unprotect() function is running everything locked. Before every
potential exit path of the function mmap_unlock() gets called to make sure
we don't leak the lock.
However, the function calls tb_invalidate_phys_page() which again can
exit a signal through longjmp, leaving our mmap_unlock() attempts in vain.
Add a hint to tb_invalidate_phys_page() that we need to unlock before we
can leave back into guest context, so that we don't leak the lock.
This fixes 16-bit i386 wine programs running in linux-user for me.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Alexander Graf [Sat, 6 Jul 2013 12:17:51 +0000 (14:17 +0200)]
linux-user: Reset copied CPUs in cpu_copy() always
When a new thread gets created, we need to reset non arch specific state to
get the new CPU into clean state.
However this reset should happen before the arch specific CPU contents get
copied over. Otherwise we end up having clean reset state in our newly created
thread.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Alexander Graf [Sat, 6 Jul 2013 12:17:53 +0000 (14:17 +0200)]
linux-user: Fix epoll on ARM hosts
The epoll emulation uses data structures without packing them, so the
compiler might choose to add padding inside.
This patch makes the most offending one (target_epoll_event) a packed
structure to make sure we don't pad it by accident. ARM would pad it,
so declare the padding mandatory for ARM targets.
This fixes i386-on-ARM epoll emulation for me.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Alexander Graf [Sat, 6 Jul 2013 12:17:49 +0000 (14:17 +0200)]
linux-user: fix segmentation fault passing with h2g(x) != x
When forwarding a segmentation fault into the guest process, we were passing
the host's address directly into the guest process's signal descriptor.
That obviously confused the guest process, since it didn't know what to make
of the (usually 32-bit truncated) address. Passing in h2g(address) makes the
guest process a lot happier.
To make the code more obvious, introduce a h2g_nocheck() macro that does the
same as h2g(), but allows us to convert addresses that may be outside of guest
mapped range into the guest's view of address space.
This fixes java running in arm-linux-user for me.
Signed-off-by: Alexander Graf <agraf@suse.de> Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Peter Maydell [Sat, 6 Jul 2013 20:44:53 +0000 (21:44 +0100)]
linux-user: Fix target_stat and target_stat64 for OpenRISC
OpenRISC uses the asm-generic versions of target_stat and
target_stat64, but it was incorrectly using the x86/ARM/etc version
due to a misplaced defined(TARGET_OPENRISC). The previously unused
OpenRISC section of the ifdef ladder also defined an incorrect
target_stat and omitted the target_stat64 definition. Fix
target_stat, provide target_stat64, and add a comment noting that
these are the asm-generic versions for the benefit of future ports.
Reviewed-by: Jia Liu <proljc@gmail.com> Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Signed-off-by: Riku Voipio <riku.voipio@linaro.org>
Older KVM version put invalid value in the segments registers dpl field for
real mode guests (0x3).
This breaks migration from those hosts to hosts with unrestricted guest support.
We detect it by checking CS dpl value for real mode guest and fix the dpl values
of all the segment registers.
Signed-off-by: Orit Wasserman <owasserm@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Older KVM versions save CS dpl value to an invalid value for real mode guests
(0x3). This patch detect this situation when loading CPU state and set all the
segments dpl to zero.
This will allow migration from older KVM on host without unrestricted guest
to hosts with restricted guest support.
For example migration from a Penryn host (with kernel 2.6.32) to
a Westmere host (for real mode guest) will fail with "kvm: unhandled exit 80000021".
Signed-off-by: Orit Wasserman <owasserm@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:58 +0000 (10:01 -0400)]
rdma: account for the time spent in MIG_STATE_SETUP through QMP
Using the previous patches, we're now able to timestamp the SETUP
state. Once we have this time, let the user know about it in the
schema.
Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:57 +0000 (10:01 -0400)]
rdma: introduce MIG_STATE_NONE and change MIG_STATE_SETUP state transition
As described in the previous patch, until now, the MIG_STATE_SETUP
state was not really a 'formal' state. It has been used as a 'zero' state
(what we're calling 'NONE' here) and QEMU has been unconditionally transitioning
into this state when the QMP migration command was called. Instead we want to
introduce MIG_STATE_NONE, which is our starting state in the state machine, and
then immediately transition into the MIG_STATE_SETUP state when the QMP migrate
command is issued.
In order to do this, we must delay the transition into MIG_STATE_ACTIVE until
later in the migration_thread(). This is done to be able to timestamp the amount of
time spent in the SETUP state for proper accounting to the user during
an RDMA migration.
Furthermore, the management software, until now, has never been aware of the
existence of the SETUP state whatsoever. This must change, because, timing of this
state implies that the state actually exists.
These two patches cannot be separated because the 'query_migrate' QMP
switch statement needs to know how to handle this new state transition.
Reviewed-by: Juan Quintela <quintela@redhat.com> Tested-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:56 +0000 (10:01 -0400)]
rdma: allow state transitions between other states besides ACTIVE
This patch is in preparation for the next ones: Until now the MIG_STATE_SETUP
state was not really a 'formal' state. It has been used as a 'zero' state
and QEMU has been unconditionally transitioning into this state when
the QMP migrate command was called. In preparation for timing this state,
we have to make this state a a 'real' state which actually gets transitioned
from later in the migration_thread() from SETUP => ACTIVE, rather than just
automatically dropping into this state at the beginninig of the migration.
This means that the state transition function (migration_finish_set_state())
needs to be capable of transitioning from valid states _other_ than just
MIG_STATE_ACTIVE.
The function is in fact already capable of doing that, but was not allowing the
old state to be a parameter specified as an input.
This patch fixes that and only makes the transition if the current state
matches the old state that the caller intended to transition from.
Reviewed-by: Juan Quintela <quintela@redhat.com> Tested-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:55 +0000 (10:01 -0400)]
rdma: send pc.ram
This takes advantages of the previous patches:
1. use the new QEMUFileOps hook 'save_page'
2. call out to the right accessor methods to invoke
the iteration hooks defined in QEMUFileOps
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Chegu Vinod <chegu_vinod@hp.com> Tested-by: Chegu Vinod <chegu_vinod@hp.com> Tested-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:54 +0000 (10:01 -0400)]
rdma: core logic
Code that does need to be visible is kept
well contained inside this file and this is the only
new additional file to the entire patch.
This file includes the entire protocol and interfaces
required to perform RDMA migration.
Also, the configure and Makefile modifications to link
this file are included.
Full documentation is in docs/rdma.txt
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Chegu Vinod <chegu_vinod@hp.com> Tested-by: Chegu Vinod <chegu_vinod@hp.com> Tested-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:53 +0000 (10:01 -0400)]
rdma: introduce ram_handle_compressed()
This gives RDMA shared access to madvise() on the destination side
when an entire chunk is found to be zero.
Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Reviewed-by: Chegu Vinod <chegu_vinod@hp.com> Tested-by: Chegu Vinod <chegu_vinod@hp.com> Tested-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Michael R. Hines [Mon, 22 Jul 2013 14:01:51 +0000 (10:01 -0400)]
rdma: update documentation to reflect new unpin support
As requested, the protocol now includes memory unpinning support.
This has been implemented in a non-optimized manner, in such a way
that one could devise an LRU or other workload-specific information
on top of the basic mechanism to influence the way unpinning happens
during runtime.
The feature is not yet user-facing, and is thus can only be enabled
at compile-time.
Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Michael R. Hines <mrhines@us.ibm.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Andreas Färber [Thu, 27 Jun 2013 17:09:09 +0000 (19:09 +0200)]
cpu: Introduce CPUClass::memory_rw_debug() for target_memory_rw_debug()
Make inline target_memory_rw_debug() always available and change its
argument to CPUState. Let it check if CPUClass::memory_rw_debug provides
a specialized callback and fall back to cpu_memory_rw_debug() otherwise.
The only overriding implementation is for 32-bit sparc.
This prepares for changing GDBState::g_cpu to CPUState.
Andreas Färber [Thu, 27 Jun 2013 11:44:40 +0000 (13:44 +0200)]
gdbstub: Change GDBState::query_cpu to CPUState
Since first_cpu/next_cpu are CPUState, CPUArchState is no longer needed.
This resolves a NULL pointer dereference of query_cpu, introduced
with commit 182735efaf956ccab50b6d74a4fed163e0f35660 and reported by
TeLeMan and Max Filippov.
Reviewed-by: Richard Henderson <rth@twiddle.net> Signed-off-by: Andreas Färber <afaerber@suse.de>
Hu Tao [Mon, 1 Jul 2013 10:18:33 +0000 (18:18 +0800)]
fw_cfg: Use QOM realize for fw_cfg
Signed-off-by: Hu Tao <hutao@cn.fujitsu.com>
[AF: Moved sysbus_init_mmio() to instance_init, renamed variable] Signed-off-by: Andreas Färber <afaerber@suse.de>
projects / mirror_qemu.git / log