Support an inter-vm shared memory device that maps a shared-memory object as a
PCI device in the guest. This patch also supports interrupts between guest by
communicating over a unix domain socket. This patch applies to the qemu-kvm
repository.
-device ivshmem,size=<size in format accepted by -m>[,shm=<shm name>]
Interrupts are supported between multiple VMs by using a shared memory server
by using a chardev socket.
-device ivshmem,size=<size in format accepted by -m>[,shm=<shm name>]
[,chardev=<id>][,msi=on][,ioeventfd=on][,vectors=n][,role=peer|master]
-chardev socket,path=<path>,id=<id>
The shared memory server, sample programs and init scripts are in a git repo here:
Andrea Arcangeli [Tue, 27 Jul 2010 19:04:36 +0000 (21:04 +0200)]
ide: Avoid canceling IDE DMA
The reason for not actually canceling the I/O is because with
virtualization and lots of VM running, a guest fs may mistake a
overload of the host, as an IDE timeout. So rather than canceling the
I/O, it's safer to wait I/O completion and simulate that the I/O has
completed just before the io cancellation was requested by the
guest. This way if ntfs or an app writes data without checking for
-EIO retval, and it thinks the write has succeeded, it's less likely
to run into troubles. Similar issues for reads.
Furthermore because the DMA operation is splitted into many synchronous
aio_read/write if there's more than one entry in the SG table, without this
patch the DMA would be cancelled in the middle, something we've no idea if it
happens on real hardware too or not. Overall this seems a great risk for zero
gain.
This approach is sure safer than previous code given we can't pretend all guest
fs code out there to check for errors and reply the DMA if it was completed
partially, given a timeout would never materialize on a real harddisk unless
there are defective blocks (and defective blocks are practically only an issue
for reads never for writes in any recent hardware as writing to blocks is the
way to fix them) or the harddisk breaks as a whole.
bdrv_eject() gets called when a device model opens or closes the tray.
If the block driver implements method bdrv_eject(), that method gets
called. Drivers host_cdrom implements it, and it opens and closes the
physical tray, and nothing else. When a device model opens, then
closes the tray, media changes only if the user actively changes the
physical media while the tray is open. This is matches how physical
hardware behaves.
If the block driver doesn't implement method bdrv_eject(), we do
something quite different: opening the tray severs the connection to
the image by calling bdrv_close(), and closing the tray does nothing.
When the device model opens, then closes the tray, media is gone,
unless the user actively inserts another one while the tray is open,
with a suitable change command in the monitor. This isn't how
physical hardware behaves. Rather inconvenient when programs
"helpfully" eject media to give you a chance to change it. The way
bdrv_eject() behaves here turns that chance into a must, which is not
what these programs or their users expect.
Change the default action not to call bdrv_close(). Instead, note the
tray status in new BlockDriverState member tray_open. Use it in
bdrv_is_inserted().
Arguably, the device models should keep track of tray status
themselves. But this is less invasive.
Kevin Wolf [Wed, 28 Jul 2010 09:26:29 +0000 (11:26 +0200)]
block: Fix bdrv_has_zero_init
Assuming that any image on a block device is not properly zero-initialized is
actually wrong: Only raw images have this problem. Any other image format
shouldn't care about it, they initialize everything properly themselves.
This patch improves the resilience of the load_vmstate() function, doing
further and better ordered tests.
In load_vmstate(), if there is any error on bdrv_snapshot_goto(), except if the
error is on VM state device, load_vmstate() will return zero and the VM will be
started with major corruption chances.
The current process:
- test if there is any writable device without snapshot support
- if exists return -error
- get the device that saves the VM state, possible return -error but unlikely
because it was tested earlier
- flush I/O
- run bdrv_snapshot_goto() on devices
- if fails, give an warning and goes to the next (not good!)
- if fails on the VM state device, return zero (not good!)
- check if the requested snapshot exists on the device that saves the VM state
and the state is not zero
- if fails return -error
- open the file with the VM state
- if fails return -error
- load the VM state
- if fails return -error
- return zero
New behavior:
- get the device that saves the VM state
- if fails return -error
- check if the requested snapshot exists on the device that saves the VM state
and the state is not zero
- if fails return -error
- test if there is any writable device without snapshot support
- if exists return -error
- test if the devices with snapshot support have the requested snapshot
- if anyone fails, return -error
- flush I/O
- run snapshot_goto() on devices
- if anyone fails, return -error
- open the file with the VM state
- if fails return -error
- load the VM state
- if fails return -error
- return zero
do_loadvm must not call vm_start if any error has occurred in load_vmstate.
Kevin Wolf [Fri, 16 Jul 2010 15:17:01 +0000 (17:17 +0200)]
block: Change bdrv_commit to handle multiple sectors at once
bdrv_commit copies the image to its backing file sector by sector, which
is (surprise!) relatively slow. Let's take a larger buffer and handle more
sectors at once if possible.
With a 1G qcow2 file, this brought the time bdrv_commit takes down from
5:06 min to 1:14 min for me.
Blue Swirl [Sat, 31 Jul 2010 19:40:17 +0000 (19:40 +0000)]
Fix uint8_t comparison with negative value
Commit 7bccf57383cca60a778d5c543ac80c9f62d89ef2 missed this one:
/src/qemu/ui/vnc-enc-tight.c: In function 'send_sub_rect':
/src/qemu/ui/vnc-enc-tight.c:1527: warning: comparison is always true due to limited range of data type
Blue Swirl [Sat, 31 Jul 2010 19:40:13 +0000 (19:40 +0000)]
Fix a warning on OpenSolaris
Add a missing #include statement to avoid a warning:
/src/qemu/net/tap-solaris.c: In function 'tap_open':
/src/qemu/net/tap-solaris.c:189: warning: implicit declaration of function 'error_report'
Amit Shah [Tue, 27 Jul 2010 10:19:19 +0000 (15:49 +0530)]
migration: Accept 'cont' only after successful incoming migration
When a 'cont' is issued on a VM that's just waiting for an incoming
migration, the VM reboots and boots into the guest, possibly corrupting
its storage since it could be shared with another VM running elsewhere.
Ensure that a VM started with '-incoming' is only run when an incoming
migration successfully completes.
A new qerror, QERR_MIGRATION_EXPECTED, is added to signal that 'cont'
failed due to no incoming migration has been attempted yet.
Joel Schopp [Wed, 21 Jul 2010 20:05:16 +0000 (15:05 -0500)]
fix variable type in qemu-io.c
The variable len can get a negative return value from cvtnum,
which we check for, but which is impossible with the current
unsigned variable type. Currently the if(len < 0) check is
pointless. This patch fixes that.
Merge branch 'for-anthony' of git://repo.or.cz/qemu/kevin
* 'for-anthony' of git://repo.or.cz/qemu/kevin:
Fix -snapshot deleting images on disk change
block: Use error codes from lower levels for error message
block: default to 0 minimal / optiomal I/O size
move 'unsafe' to end of caching modes in help
virtio-blk: Create exit function to unregister savevm
block migration: propagate return value when bdrv_write() returns < 0
ide/atapi: add support for GET EVENT STATUS NOTIFICATION
linux-user: fix build on hosts not using guest base
Commit 68a1c816868b3e35a1da698af412b29e61b1948a broke qemu on hosts not
using guest base. It uses reserved_va unconditionally in mmap.c. To
avoid to many #ifdef #endif blocks, define RESERVED_VA as either
reserved_va or 0ul, and use it instead of reserved_va, similarly to what
has been done with guest_base/GUEST_BASE.
linux-user: Re-use load_elf_image for the main binary.
This requires moving the PT_INTERP extraction and GUEST_BASE
handling into load_elf_image. Key this off a non-null pointer
argument to receive the interpreter name.
linux-user: Extract load_elf_image from load_elf_interp.
Moving toward a single copy of the elf binary loading code.
Fill in the details of the loaded image into a struct image_info.
Adjust create_elf_tables to read from such structures instead
of from a collection of passed arguments. Don't return error
values from load_elf_interp; always exit(-1) with a message to
stderr. Collect elf_interpreter handling in load_elf_binary
to a common spot.
linux-user: Remove partial support for a.out interpreters.
At the bottom of the a.out support was the unimplemented load_aout_interp
function. There were other portions of the support that didn't look
right; when I went to look in the Linux kernel for clarification, I found
that the support for such interpreters has been removed from binfmt_elf.
There doesn't seem to be any reason to keep this broken support in qemu.
There are no supported stack-grows-up targets. We were putting
the guard page at the highest address, i.e. the bottom of the stack.
Use the maximum of host and guest page size for the guard size.
First, adjust load_symbols to accept a load_bias parameter. At the same
time, read the entire section header table in one go, use pread instead
f lseek+read for the symbol and string tables, and properly free
allocated structures on error exit paths.
Second, adjust load_elf_interp to compute load_bias. This requires
finding out the built-in load addresses. Which allows us to honor a
pre-linked interpreter image when possible, and eliminate the hard-coded
INTERP_MAP_SIZE value.
Remove ifdefs from code by defining empty inline functions
when byte swapping isn't needed. Push loops over swapping
arrays of structures into the swapping functions.
linux-user: Reduce lseek+reads while loading elf files.
Define BPRM_BUF_SIZE to 1k and read that amount initially. If the
data we want from the binary is in this buffer, use it instead of
reading from the file again.
linux-user: Handle filesz < memsz for any PT_LOAD segment.
I caught padzero not properly initializing the .bss segment
on a statically linked Alpha program. Rather than a minimal
patch, replace the gross code with a single mmap+memset.
Share more code between load_elf_interp and load_elf_binary.
Legally, an ELF program need not have just a single .bss;
and PT_LOAD segment can have memsz > filesz.
Blue Swirl [Tue, 27 Jul 2010 17:26:08 +0000 (17:26 +0000)]
Fix mingw32 build
Fix mingw32 build errors like
/src/qemu/ui/vnc-enc-tight.c: In function 'tight_detect_smooth_image24':
/src/qemu/ui/vnc-enc-tight.c:119: error: 'uint' undeclared (first use in this function)
Blue Swirl [Tue, 27 Jul 2010 15:34:10 +0000 (15:34 +0000)]
Initialize a variable in all cases
Fix a warning with some GCCs:
/src/qemu/ui/vnc-enc-tight.c: In function `send_sub_rect_nojpeg':
/src/qemu/ui/vnc-enc-tight.c:1458: warning: `ret' might be used uninitialized in this function
Blue Swirl [Tue, 27 Jul 2010 15:32:39 +0000 (15:32 +0000)]
Fix uint8_t comparisons with negative values
Fix the following warnings:
/src/qemu/hw/ide/core.c: In function `ide_drive_pio_post_load':
/src/qemu/hw/ide/core.c:2767: warning: comparison is always false due to limited range of data type
/src/qemu/ui/vnc-enc-tight.c: In function `tight_detect_smooth_image':
/src/qemu/ui/vnc-enc-tight.c:284: warning: comparison is always true due to limited range of data type
/src/qemu/ui/vnc-enc-tight.c:297: warning: comparison is always true due to limited range of data type
/src/qemu/ui/vnc-enc-tight.c: In function `tight_encode_indexed_rect16':
/src/qemu/ui/vnc-enc-tight.c:456: warning: comparison is always false due to limited range of data type
/src/qemu/ui/vnc-enc-tight.c: In function `tight_encode_indexed_rect32':
/src/qemu/ui/vnc-enc-tight.c:457: warning: comparison is always false due to limited range of data type
Implement a threaded VNC server using the producer-consumer model.
The main thread will push encoding jobs (a list a rectangles to update)
in a queue, and the VNC worker thread will consume that queue and send
framebuffer updates to the output buffer.
The threaded VNC server can be enabled with ./configure --enable-vnc-thread.
If you don't want it, just use ./configure --disable-vnc-thread and a syncrhonous
queue of job will be used (which as exactly the same behavior as the old queue).
If you disable the VNC thread, all thread related code will not be built and there will
be no overhead.
qemu-thread: add qemu_mutex/cond_destroy and qemu_mutex_exit
Add some missing functions in qemu-thread. Currently qemu-thread
is only used for io-thread but it will used by the vnc server soon
and we need those functions instead of calling pthread directly.
Profiling with callgrind seems to show that a lot of time is spent
in the palette code (mostly due to memory allocation and qdict to int
conversion).
This patch adds a VncPalette implementation. The palette is stored
in a hash table, like qdict, but which does way less memory allocations,
and doesn't suffer from the QObject overhead.
vnc: tight: specific zlib level and filters for each compression level
Disable png filters for lower compression levels. This should lower
the CPU consumption and reduce encoding time.
This isn't in tight_conf because:
* tight_conf structure must not change, because it's shared with other
tight implementations (libvncserver, etc..).
* it'd exceed the 80 col limit.
* PNG_ macros are only defined if CONFIG_VNC_PNG is defined
Introduce a new encoding: VNC_ENCODING_TIGHT_PNG [1] (-269) with a new
tight filter VNC_TIGHT_PNG (0x0A). When the client tells it supports the Tight PNG
encoding, the server will use tight, but will always send encoding pixels using
PNG instead of zlib. If the client also told it support JPEG, then the server can
send JPEG, because PNG will only be used in the cases zlib was used in normal tight.
This encoding was introduced to speed up HTML5 based VNC clients like noVNC [2], but
can also be used on devices like iPhone where PNG can be rendered in hardware.
vnc: tight: add JPEG and gradient subencoding with smooth image detection
Add gradient filter and JPEG compression with an heuristic to detect how
lossy the comppression will be. This code has been adapted from
libvncserver/tight.c.
JPEG support can be enabled/disabled at compile time with --enable-vnc-jpeg
and --disable-vnc-jpeg.
Juan Quintela [Mon, 26 Jul 2010 19:38:43 +0000 (21:38 +0200)]
vmstate: add subsections code
This commit adds subsections for each device section.
Subsections is the way to handle information that don't need to be sent
to de destination of a migration because its values are not needed. It is
the way to handle optional information. Notice that only the source can
decide if the information is optional or not. The destination needs to
understand all subsections that it receives to have a sucessful load.
The conflicts are due to commit 4fc8d6711aff7a9c11e402c3d77b481609f9f486
that is a fix to the ide_drive_pre_save() function. It reverts both
(and both are reinstantiated later in the series)
Currently we set them to 512 bytes unless manually specified. Unforuntaly
some brain-dead partitioning tools create unaligned partitions if they
get low enough optiomal I/O size values, so don't report any at all
unless explicitly set.
Bruce Rogers [Wed, 21 Jul 2010 20:32:28 +0000 (14:32 -0600)]
move 'unsafe' to end of caching modes in help
Libvirt parses qemu help output to determine qemu features. In particular
it probes for the following: "cache=writethrough|writeback|none". The
addition of the unsafe cache mode was inserted within this string, as
opposed to being added to the end, which impacted libvirt's probe.
Unbreak libvirt by keeping the existing cache modes intact and add
unsafe to the end.
This problem only manifests itself if a caching mode is explicitly
specified in the libvirt xml, in which case older syntax for caching is
passed to qemu, which it no longer understands.
block migration: propagate return value when bdrv_write() returns < 0
Currently block_load() doesn't check return value of bdrv_write(), and
even the destination weren't prepared to execute block migration, it
proceeds and guest boots on the target. This patch fix this issue.
ide/atapi: add support for GET EVENT STATUS NOTIFICATION
The GET EVENT STATUS NOTIFICATION is a mandatory command according
to MMC-3, even if event status notification is not supported.
This patch adds support for this command. It returns NEA ("No Event
Available") with an empty "Supported Event Classes" to show that it
doesn't event support status notification. If asychronous operation is
requested, which requires NCQ support, it returns an error according
to the specifications.
This fixes HAL support on FreeBSD and derivatives, which fill up the
logs every second with:
Stefan Weil [Thu, 22 Jul 2010 20:15:24 +0000 (22:15 +0200)]
slirp: Remove declarations which are no longer needed
The previous patches replaced u_int8_t, u_int16_t, u_int32_t, u_int64_t
by standard int types from stdint.h,
so we can now remove their declarations which are no longer needed.
projects / qemu.git / log