Paul Brook [Fri, 12 Mar 2010 16:54:58 +0000 (16:54 +0000)]
Add tb_page_addr_t
The page tracking code in exec.c is used by both userspace and system
emulation. Userspace emulation uses it to track virtual pages, and
system emulation to track ram pages. Introduce a new type to hold this
kind of address.
Define L1_MAP_ADDR_SPACE_BITS to be either the virtual address size
(in user mode) or physical address size (in system mode), and use
that to size l1_map. This rewrites page_find_alloc, page_flush_tb,
and walk_memory_regions.
Use TARGET_PHYS_ADDR_SPACE_BITS for the physical memory map based
off of l1_phys_map. This rewrites page_phys_find_alloc and
phys_page_for_each.
Don't return addresses that aren't properly aligned for the guest,
e.g. when the guest has a larger page size than the host. Don't
return addresses that are outside the virtual address space for the
target, by paying proper attention to the h2g/g2h macros.
At the same time, place the default mapping base for 64-bit guests
(on 64-bit hosts) outside the low 4G. Consistently interpret
mmap_next_start in the guest address space.
Previously, only 32-bit guests had a proper check for the
validity of the virtual address. Extend that check to 64-bit
guests with a restricted virtual address space.
Move TARGET_PHYS_ADDR_SPACE_BITS to target-*/cpu.h.
Removes a set of ifdefs from exec.c.
Introduce TARGET_VIRT_ADDR_SPACE_BITS for all targets other
than Alpha. This will be used for page_find_alloc, which is
supposed to be using virtual addresses in the first place.
Aurelien Jarno [Sat, 6 Mar 2010 17:02:31 +0000 (18:02 +0100)]
target-i386: fix SIB decoding with index = 4
A SIB byte with an index of 4 means "no scaled index", even if the scale
value is not 0. In 64-bit mode, if REX.X is used, an index of 4 selects
%r12. This is correctly handled by the computation of the index variable,
which includes the index bits, and also the REX.X prefix:
index = ((code >> 3) & 7) | REX_X(s);
Thanks to Avi Kivity, Jamie Lokier and Malc for the analysis of the
problem and the initial patch.
Shahar Havivi [Fri, 26 Feb 2010 09:34:59 +0000 (11:34 +0200)]
Restore terminal attributes for tty based monitor
Patch http://permalink.gmane.org/gmane.comp.emulators.qemu/63472 handle
close when using tty devices (like /dev/ttyS0),
yet tty based monitor are not restoring terminal attributes (as done
with stdio based monitor), when closing qemu after that command:
$ qemu -monitor /dev/tty
the terminal is not responding until you write reset (blindly),
this patch fix it
Gerd Hoffmann [Fri, 26 Feb 2010 16:17:39 +0000 (17:17 +0100)]
kbd keds: vnc
Use led status notification support in vnc.
The qemu vnc server keeps track of the capslock and numlock states based
on the key presses it receives from the vnc client. But this fails in
case the guests idea of the capslock and numlock state changes for other
reasons. One case is guest reboot (+ keyboard reset). Another case are
more recent windows versions which reset capslock state before
presenting the login screen.
Usually guests use the keyboard leds to signal the capslock and numlock
state to the user, so we can use this to better keep track of capslock
and numlock state in the qemu vnc server.
Also toggle the numlock and capslock states on keydown events (instead
of keyup). Guests do the same.
qdev: Catch attempt to attach more than one device to a netdev
Guest device and host netdev are peers, i.e. it's a 1:1 relation.
However, we fail to enforce that:
$ qemu -nodefaults --nographic -netdev user,id=net0 -device e1000,netdev=net0 -device virtio-net-pci,netdev=net0 -monitor stdio
QEMU 0.12.50 monitor - type 'help' for more information
(qemu) info network
Devices not on any VLAN:
net0: net=10.0.2.0, restricted=n peer=virtio-net-pci.0
e1000.0: model=e1000,macaddr=52:54:00:12:34:56 peer=net0
virtio-net-pci.0: model=virtio-net-pci,macaddr=52:54:00:12:34:57 peer=net0
Luiz Capitulino [Thu, 25 Feb 2010 15:07:01 +0000 (12:07 -0300)]
QMP: Drop DEBUG event
This event has been introduced in the first round of QMP commits,
turns out that it's based on the usage of the EXCP_DEBUG macro,
which has discussable semantics when exposed through QMP.
Luiz Capitulino [Thu, 25 Feb 2010 15:07:00 +0000 (12:07 -0300)]
QMP: Move RESET event into qemu_system_reset()
Nothing will change as that function is currently only called by
the main loop code, but it's the right place for the RESET event,
as it's where the reset is actually performed.
Luiz Capitulino [Thu, 25 Feb 2010 15:06:58 +0000 (12:06 -0300)]
block: Emit BLOCK_IO_ERROR before vm_stop() call
The next commit will move the STOP event into do_vm_stop(), to
have the expected event sequence we need to emit the I/O error
event before calling vm_stop().
Adam Litke [Mon, 22 Feb 2010 16:51:20 +0000 (10:51 -0600)]
Fix hanging user monitor when using balloon command
This patch application failed. My patch adds a cb() call in
do_balloon(), but the change in git has added the cb() call to
do_info_balloon(). That is causing qemu segfaults. Applying the
following should correct the damage. Thanks.
Samuel Thibault [Sun, 28 Feb 2010 20:03:00 +0000 (21:03 +0100)]
Fix curses interaction with keymaps
The combination of keymap support (-k option) and curses is currently
very broken. The patch below fixes it by first extending keymap support
to interpret the shift, ctrl, altgr and addupper keywords in keymaps,
and to fix curses into properly using keymaps.
Stefan Weil [Thu, 4 Mar 2010 21:21:02 +0000 (22:21 +0100)]
Documentation: Modify rule for html output (better looking output format)
To create html output from texi input, texi2html was used.
Output from makeinfo looks cleaner, so replace the old rule
and use makeinfo now.
For those who want to use their own variant of html output,
the macros MAKEINFO and MAKEINFOFLAGS allow customisation.
Option "-I ." is not needed (the current directory is
searched by default), so remove it.
Please note that the build requirements changed, too:
makeinfo is required for doc builds.
texi2html is no longer used.
Kevin Wolf [Wed, 24 Feb 2010 15:17:58 +0000 (16:17 +0100)]
json-parser: Fix segfault on malformed input
If the parser fails to parse the key in parse_pair, it will access a NULL
pointer. A simple way to trigger this is sending {foo} via QMP. This patch
turns the segfault into a syntax error reply.
Jan Kiszka [Mon, 1 Mar 2010 18:10:32 +0000 (19:10 +0100)]
x86: Extend validity of bsp_to_cpu
As we hard-wire the BSP to CPU 0 anyway and cpuid_apic_id equals
cpu_index, bsp_to_cpu can also be based on the latter directly. This
will help an early user of it: KVM while initializing mp_state.
Jan Kiszka [Mon, 1 Mar 2010 18:10:31 +0000 (19:10 +0100)]
KVM: x86: Restrict writeback of VCPU state
Do not write nmi_pending, sipi_vector, and mpstate unless we at least go
through a reset. And TSC as well as KVM wallclocks should only be
written on full sync, otherwise we risk to drop some time on state
read-modify-write.
Jan Kiszka [Mon, 1 Mar 2010 18:10:30 +0000 (19:10 +0100)]
KVM: Rework VCPU state writeback API
This grand cleanup drops all reset and vmsave/load related
synchronization points in favor of four(!) generic hooks:
- cpu_synchronize_all_states in qemu_savevm_state_complete
(initial sync from kernel before vmsave)
- cpu_synchronize_all_post_init in qemu_loadvm_state
(writeback after vmload)
- cpu_synchronize_all_post_init in main after machine init
- cpu_synchronize_all_post_reset in qemu_system_reset
(writeback after system reset)
These writeback points + the existing one of VCPU exec after
cpu_synchronize_state map on three levels of writeback:
- KVM_PUT_RUNTIME_STATE (during runtime, other VCPUs continue to run)
- KVM_PUT_RESET_STATE (on synchronous system reset, all VCPUs stopped)
- KVM_PUT_FULL_STATE (on init or vmload, all VCPUs stopped as well)
This level is passed to the arch-specific VCPU state writing function
that will decide which concrete substates need to be written. That way,
no writer of load, save or reset functions that interact with in-kernel
KVM states will ever have to worry about synchronization again. That
also means that a lot of reasons for races, segfaults and deadlocks are
eliminated.
cpu_synchronize_state remains untouched, just as Anthony suggested. We
continue to need it before reading or writing of VCPU states that are
also tracked by in-kernel KVM subsystems.
Consequently, this patch removes many cpu_synchronize_state calls that
are now redundant, just like remaining explicit register syncs.
Jan Kiszka [Mon, 1 Mar 2010 18:10:29 +0000 (19:10 +0100)]
KVM: Rework of guest debug state writing
So far we synchronized any dirty VCPU state back into the kernel before
updating the guest debug state. This was a tribute to a deficite in x86
kernels before 2.6.33. But as this is an arch-dependent issue, it is
better handle in the x86 part of KVM and remove the writeback point for
generic code. This also avoids overwriting the flushed state later on if
user space decides to change some more registers before resuming the
guest.
We furthermore need to reinject guest exceptions via the appropriate
mechanism. That is KVM_SET_GUEST_DEBUG for older kernels and
KVM_SET_VCPU_EVENTS for recent ones. Using both mechanisms at the same
time will cause state corruptions.
Avi Kivity [Wed, 24 Feb 2010 21:11:19 +0000 (18:11 -0300)]
Allocate memory below 4GB as one chunk
Instead of allocating a separate chunk for the first 640KB and another
for 1MB+, allocate one large chunk. This plays well in terms of alignment
and size with large pages.
Stefan Weil [Tue, 2 Mar 2010 21:37:57 +0000 (22:37 +0100)]
eepro100: New function for reading command block
Move code which reads the command block to the
new function read_cb. The patch also fixes some
endianess issues related to the command block
and moves declarations of local variables to
the beginning of the block.
projects / qemu.git / log