Blue Swirl [Tue, 1 Nov 2011 20:57:01 +0000 (20:57 +0000)]
Merge branch 'ppc-next' of git://repo.or.cz/qemu/agraf
* 'ppc-next' of git://repo.or.cz/qemu/agraf: (24 commits)
pseries: Add partial support for PCI
ppc: Alter CPU state to mask out TCG unimplemented instructions as appropriate
pseries: Allow writes to KVM accelerated TCE table
KVM: PPC: Override host vmx/vsx/dfp only when information known
ppc: Fix up usermode only builds
pseries: Correct vmx/dfp handling in both KVM and TCG cases
PPC: Fail configure when libfdt is not available
ppc: Avoid decrementer related kvm exits
PPC: Disable non-440 CPUs for ppcemb target
PPC: Bump qemu-system-ppc to 64-bit physical address space
pseries: Under kvm use guest cpu = host cpu by default
ppc: Add cpu defs for POWER7 revisions 2.1 and 2.3
ppc: First cut implementation of -cpu host
ppc: Remove broken partial PVR matching
pseries: Update SLOF firmware image
pseries: Add device tree properties for VMX/VSX and DFP under kvm
ppc: Generalize the kvmppc_get_clockfreq() function
Set an invalid-bits mask for each SPE instructions
pseries: Update SLOF firmware image
pseries: Use Book3S-HV TCE acceleration capabilities
...
Max Filippov [Thu, 27 Oct 2011 22:10:27 +0000 (02:10 +0400)]
opencores_eth: fix RX path: FCS, padding and TL
OpenCores 10/100 Mbps Ethernet MAC specification doesn't clearly state
whether FCS is counted in the RX frame length or not. Looks like it is.
Append zero FCS to the received frames.
Get rid of big static buffer for RX frame padding, optimize it for the
most common MINFL value range.
Set RXD_TL for the long frames only when HUGEN bit is set.
Gerd Hoffmann [Thu, 27 Oct 2011 07:12:04 +0000 (09:12 +0200)]
migration: flush migration data to disk.
This patch increases robustness when migrating to a file with
two little changes:
(1) Before closing the migration file handle checks if it happens to be
a regular file and if so it issues a fsync. This way the data is
flushed to disk before qemu sends the migration completed event.
(2) It adds error checking. In case either fsync or close syscall
fails pass up the error (and fail migration).
Luiz Capitulino [Fri, 28 Oct 2011 16:59:52 +0000 (14:59 -0200)]
Fix segfault on migration completion
A simple migration reproduces it:
1. Start the source VM with:
# qemu [...] -S
2. Start the destination VM with:
# qemu <source VM cmd-line> -incoming tcp:0:4444
3. In the source VM:
(qemu) migrate -d tcp:0:4444
4. The source VM will segfault as soon as migration completes (might not
happen in the first try)
What is happening here is that qemu_file_put_notify() can end up closing
's->file' (in which case it's also set to NULL). The call stack is rather
complex, but Eduardo helped tracking it to:
David Gibson [Fri, 28 Oct 2011 04:26:18 +0000 (15:26 +1100)]
Add qapi related generated files to .gitignore
.gitignore already lists the qapi-generated subdirectory which includes a
number of files generated during build. However, there are some additional
files generated by the qapi build which go in the top level directory.
This patch adds them to .gitignore, removing the irritating noise from
diffs and the like.
Eric Sunshine [Tue, 1 Nov 2011 09:39:49 +0000 (05:39 -0400)]
qemu-barrier: Fix build failure on PowerPC Mac OS X
qemu-barrier.h tests if macro __powerpc__ is defined, however, the
preprocessor on PowerPC Mac OS X defines only __POWERPC__, not
__powerpc__. Resolve by testing instead for qemu-provided _ARCH_PPC.
Peter Maydell [Fri, 28 Oct 2011 09:52:40 +0000 (10:52 +0100)]
qemu-tls.h: Add abstraction layer for TLS variables
Add an abstraction layer for defining and using thread-local
variables. For the moment this is implemented only for Linux,
which means they can only be used in restricted circumstances.
The abstraction layer allows us to add POSIX and Win32 support
later.
Alex Williamson [Mon, 31 Oct 2011 14:54:09 +0000 (08:54 -0600)]
Error check find_ram_offset
Spotted via code review, we initialize offset to 0 to avoid a
compiler warning, but in the unlikely case that offset is
never set to something else, we should abort instead of return
a value that will almost certainly cause problems.
Gerd Hoffmann [Fri, 28 Oct 2011 14:13:50 +0000 (16:13 +0200)]
usb-host: fix host close
The whole usb_host_close() function is skipped in case the device is not
in attached state. This is wrong though, only then usb_device_detach()
must be skipped, all other cleanup (especially device reset and closing
the file handle) still needs to be done. There are code paths where
usb_host_close() is called with the device in detached state already.
This fixes usb-host devices not being released and returned to the host
after removing them with device_del.
Roy Tam [Thu, 15 Sep 2011 03:25:47 +0000 (11:25 +0800)]
usb: change VID/PID for usb-hub and usb-msd to prevent conflict
Some USB drivers, for example USBASPI.SYS, will skip different type of
device which has same VID/PID. The following patch helps preventing
usb-msd being skipped by the driver.
Stefan Weil [Mon, 31 Oct 2011 20:29:46 +0000 (21:29 +0100)]
Support running QEMU on Valgrind
Valgrind is a tool which can automatically detect many kinds of bugs.
Running QEMU on Valgrind with x86_64 hosts was not possible because
Valgrind aborts when memalign is called with an alignment larger than
1 MiB. QEMU normally uses 2 MiB on Linux x86_64.
Now the alignment is reduced to the page size when QEMU is running on
Valgrind.
v2:
Instead of using the macro RUNNING_ON_VALGRIND from valgrind.h,
the patch now uses a hack from libvirt which tests for the pre-loaded
vgpreload_*.so shared libraries. This avoids the need for valgrind.h.
Mark Wu [Mon, 31 Oct 2011 03:29:04 +0000 (11:29 +0800)]
trace: Add wildcard trace event support
A basic wildcard matching is supported in both the monitor command
"trace-event" and the events list file. That means you can enable/disable
the events having a common prefix in a batch. For example, virtio-blk trace
events could be enabled using:
trace-event virtio_blk_* on
Gerd Hoffmann [Fri, 14 Oct 2011 09:36:43 +0000 (11:36 +0200)]
usb-hub: wakeup on attach
When attaching a new device we must send a wakeup request to the root
hub, otherwise the guest will not notice the new device in case the
usb hub is suspended.
Alon Levy [Tue, 25 Oct 2011 13:39:50 +0000 (15:39 +0200)]
qxl: create slots on post_load in vga state
RHBZ 740547
If we migrate when the device is in vga state the guest
still believes the slots are created, and will cause operations
that reference the slots, causing a "panic: virtual address out of range"
on the first of them. Easy to see by migrating in vga mode with
a driver loaded, for instance windows cmd window in full screen mode,
and then exiting vga mode back to native mode will cause said panic.
Fixed by doing the slot recreation in post_load for vga mode as well.
Note that compat does not require any changes because it creates it's
only slot by a side effect of QXL_IO_SET_MODE.
Gerd Hoffmann [Fri, 21 Oct 2011 13:59:07 +0000 (15:59 +0200)]
qxl: stride fixup
spice uses negative stride value to signal the bitmap is upside down.
The qxl renderer (used for scl, vnc and screenshots) wants a positive
value because it is easier to work with. The positive value is then
stored in the very same variable, which has the drawback that the
upside-down test works only once. Fix by using two variables.
$ MALLOC_PERTURB_=234 qemu-system-x86_64 -vnc :0,acl,sasl [...]
QEMU 0.15.50 monitor - type 'help' for more information
(qemu) acl_add vnc.username fred allow
acl: added rule at position 1
(qemu) acl_reset vnc.username
Segmentation fault (core dumped)
C99 7.15.1: Each invocation of the va_start and va_copy macros shall
be matched by a corresponding invocation of the va_end macro in the
same function.
Spotted by Coverity. Harmless on the (common) systems where va_end()
does nothing.
Stefan Weil [Wed, 5 Oct 2011 18:03:02 +0000 (20:03 +0200)]
tcg: Add bytecode generator for tcg interpreter
Unlike other tcg target code generators, this one does not generate
machine code for some cpu. It generates machine independent bytecode
which is interpreted later.
This allows running QEMU on any host.
Interpreted bytecode is slower than direct execution of generated
machine code.
qemu-queue: Introduce QLIST_INSERT_HEAD_RCU and dummy RCU wrappers.
SynthFS needs a QLIST_INSERT_HEAD_RCU to make sure list instructions are not
re-ordered and therefore avoiding a crash. There may be parallel readers which
should be allowed for lock-free access and this variant allows us to get rid
of rwlocks used by readers.
SynthFS is a special case where we dont really need full RCU capabilities as
it doesnt allow list entry deletion but concurrent readers/writers and
instruction re-ordering should not result in a crash.
Also, once the real rcu is available, dummy rcu macro definitions will go away
and the code will still work as expected.
This patchwork is based on inputs from Paolo Bonzini.
hw/9pfs: Add synthetic file system support using 9p
This patch create a synthetic file system with mount tag
v_synth when -virtfs_synth command line option is specified
in qemu. The synthetic file system can be mounted in guest
using 9p using the below command line
mount -t 9p -oversion=9p2000.L,trans=virtio v_synth <mountpint>
Synthetic file system enabled different qemu subsystem to register
callbacks for read and write events from guest. The subsystem
can create directories and files in the synthetic file system as show
in ex below
M. Mohan Kumar [Tue, 25 Oct 2011 06:40:39 +0000 (12:10 +0530)]
hw/9pfs: Read-only support for 9p export
A new fsdev parameter "readonly" is introduced to control accessing 9p export.
"readonly" can be used to specify the access type. By default "rw" access
is given to 9p export.
configure: Update configure so that open_by_handle_at check returns correct value
According to David Gibson for some compiler/libc combinations, open_by_handle_at
test in configure isn't quite right: because the file_handle pointer is never
dereferenced, gcc doesn't complain even if it is undefined. Change the test
as suggested by him.
Mathieu Sonet [Fri, 28 Oct 2011 09:55:37 +0000 (10:55 +0100)]
Add AACI audio playback support to the ARM Versatile/PB platform
This driver emulates the ARM AACI interface (PL041) connected to a LM4549 codec.
It enables audio playback for the Versatile/PB platform.
Limitations:
- Supports only a playback on one channel (Versatile/Vexpress)
- Supports only one TX FIFO in compact-mode or non-compact mode.
- Supports playback of 12, 16, 18 and 20 bits samples.
- Record is not supported.
- The PL041 is hardwired to a LM4549 codec.
Versatile/PB test build:
linux-2.6.38.5
buildroot-2010.11
alsa-lib-1.0.22
alsa-utils-1.0.22
mpg123-0.66
Qemu host: Ubuntu 10.04 in Vmware/OS X
Playback tested successfully with speaker-test/aplay/mpg123.
Signed-off-by: Mathieu Sonet <[email protected]>
[Peter Maydell: fixed typo in code clearing SL1RXBUSY/SL2RXBUSY
bits, as spotted by Andrzej Zaborowski] Signed-off-by: Peter Maydell <[email protected]> Signed-off-by: Andrzej Zaborowski <[email protected]>
David Gibson [Sun, 30 Oct 2011 17:16:46 +0000 (17:16 +0000)]
pseries: Add partial support for PCI
This patch adds a PCI bus to the pseries machine. This instantiates
the qemu generic PCI bus code, advertises a PCI host bridge in the
guest's device tree and implements the RTAS methods specified by PAPR
to access PCI config space. It also sets up the memory regions we
need to provide windows into the PCI memory and IO space, and
advertises those to the guest.
However, because qemu can't yet emulate an IOMMU, which is mandatory on
pseries, PCI devices which use DMA (i.e. most of them) will not work with
this code alone. Still, this is enough to support the virtio_pci device
(which probably _should_ use emulated PCI DMA, but is specced to use
direct hypervisor access to guest physical memory instead).
[agraf] remove typedef which could cause compile errors
David Gibson [Sun, 30 Oct 2011 15:51:24 +0000 (15:51 +0000)]
ppc: Alter CPU state to mask out TCG unimplemented instructions as appropriate
The CPU state contains two bitmaps, initialized from the CPU spec
which describes which instructions are implemented on the CPU. A
couple of bits are defined which cover instructions (VSX and DFP)
which are not currently implemented in TCG. So far, these are only
used to handle the case of -cpu host because a KVM guest can use
the instructions when the host CPU supports them.
However, it's a mild layering violation to simply not include those
bits in the CPU descriptions for those CPUs that do support them,
just because we can't handle them in TCG. This patch corrects the
situation, so that the instruction bits _are_ shown correctly in the
cpu spec table, but are masked out from the cpu state in the non-KVM
case.
David Gibson [Thu, 27 Oct 2011 15:56:31 +0000 (15:56 +0000)]
pseries: Allow writes to KVM accelerated TCE table
Sufficiently recent kernels include a KVM call to accelerate use of
PAPR TCE tables (IOMMU), which are used by PAPR virtual IO devices.
This involves qemu mapping the TCE table in from a kernel obtained fd,
which currently we do with PROT_READ only. This is a hangover from
early (never released) versions of this kernel interface which only
permitted read-only mappings and required us to destroy and recreate
the table when we needed to clear it from qemu.
Now, the kernel permits read-write mappings, and we rely on this to
clear the table in spapr_vio_quiesce_one(). However, due to
insufficient testing, I forgot to update the actual mapping of the
table in kvmppc_create_spapr_tce() to add PROT_WRITE to the mmap().
Alexander Graf [Mon, 24 Oct 2011 18:43:22 +0000 (20:43 +0200)]
KVM: PPC: Override host vmx/vsx/dfp only when information known
The -cpu host feature tries to find out the host capabilities based
on device tree information. However, we don't always have that available
because it's an optional property in dt.
So instead of force unsetting values depending on an unreliable source
of information, let's just try to be clever about it and not override
capabilities when we don't know the device tree pieces.
This fixes altivec with -cpu host on YDL PowerStations.
David Gibson [Mon, 17 Oct 2011 18:15:41 +0000 (18:15 +0000)]
pseries: Correct vmx/dfp handling in both KVM and TCG cases
Currently, when KVM is enabled, the pseries machine checks if the host
CPU supports VMX, VSX and/or DFP instructions and advertises
accordingly in the guest device tree. It does this regardless of what
CPU is selected on the command line. On the other hand, when in TCG
mode, it never advertises any of these facilities, even basic VMX
(Altivec) which is supported in TCG.
Now that we have a -cpu host option for ppc, it is fairly
straightforward to fix both problems. This patch changes the -cpu
host code to override the basic cpu spec derived from the PVR with
information queried from the host avout VMX, VSX and DFP capability.
The pseries code then uses the instruction availability advertised in
the cpu state to set the guest device tree correctly for both the KVM
and TCG cases.
Alexander Graf [Tue, 18 Oct 2011 00:10:36 +0000 (02:10 +0200)]
PPC: Fail configure when libfdt is not available
We have several targets in the PPC tree now that basically require libfdt
to function properly, namely the pseries and the e500 targets. This dependency
will rather increase than decrease in the future, so I want to make sure
that people building shiny new 1.0 actually have libfdt installed to get
rid of a few ifdefs in the code.
Warning: This patch will likely make configure fail for people who don't
select their own --target-list, but don't have libfdt development packages
installed. However, we really need this new dependency to move on.
David Gibson [Sun, 16 Oct 2011 19:26:17 +0000 (19:26 +0000)]
ppc: Avoid decrementer related kvm exits
In __cpu_ppc_store_decr(), we set up a regular timer used to trigger
decrementer interrupts. This is necessary to implement the decrementer
properly under TCG, but is unnecessary under KVM (true for both Book3S-PR
and Book3S-HV KVM variants), because the kernel handles generating and
delivering decrementer exceptions.
Under kvm, in fact, the timer causes expensive and unnecessary exits from
kvm to qemu. This patch, therefore, disables setting the timer when kvm
is in use.
Alexander Graf [Tue, 18 Oct 2011 00:00:51 +0000 (02:00 +0200)]
PPC: Disable non-440 CPUs for ppcemb target
The sole reason we have the ppcemb target is to support MMUs that have
less than the usual 4k possible page size. There are very few of these
chips and I don't want to add additional QA and testing burden to everyone
to ensure that code still works when TARGET_PAGE_SIZE is not 4k.
So this patch disables all CPUs except for MMU_BOOKE capable ones from
the ppcemb target.
David Gibson [Wed, 12 Oct 2011 22:40:34 +0000 (22:40 +0000)]
pseries: Under kvm use guest cpu = host cpu by default
Now that we've implemented -cpu host for ppc, this patch updates the
pseries machine to use the host cpu as the guest cpu by default when
running under KVM. This is important because under KVM Book3S-HV the guest
cpu _cannot_ be of a different type to the host cpu (at the moment
KVM Book3S-HV will silently virtualize the host cpu instead of whatever was
requested, but in future it is likely to simply refuse to run the VM if
a cpu model other than the host's is requested).
David Gibson [Wed, 12 Oct 2011 22:40:32 +0000 (22:40 +0000)]
ppc: First cut implementation of -cpu host
For convenience with kvm, x86 allows the user to specify -cpu host on the
qemu command line, which means make the guest cpu the same as the host
cpu. This patch implements the same option for ppc targets.
For now, this just read the host PVR (Processor Version Register) and
selects one of our existing CPU specs based on it. This means that the
option will not work if the host cpu is not supported by TCG, even if that
wouldn't matter for use under kvm.
In future, we can extend this in future to override parts of the cpu spec
based on information obtained from the host (via /proc/cpuinfo, the host
device tree, or explicit KVM calls). That will let us handle cases where
the real kvm-virtualized CPU doesn't behave exactly like the TCG-emulated
CPU. With appropriate annotation of the CPU specs we'll also then be able
to use host cpus under kvm even when there isn't a matching full TCG model.
David Gibson [Wed, 12 Oct 2011 22:40:31 +0000 (22:40 +0000)]
ppc: Remove broken partial PVR matching
The ppc target contains a ppc_find_by_pvr() function, which looks up a
CPU spec based on a PVR (that is, based on the value in the target cpu's
Processor Version Register). PVR values contain information on both the
cpu model (upper 16 bits, usually) and on the precise revision (low 16
bits, usually).
ppc_find_by_pvr, as well as making exact PVR matches, attempts to find
"close" PVR matches, when we don't have a CPU spec for the exact revision
specified. This sounds like a good idea, execpt that the current logic
is completely nonsensical.
It seems to assume CPU families are subdivided bit by bit in the PVR in a
way they just aren't. Specifically, it requires a match on all bits of the
specified pvr up to the last non-zero bit. This has the bizarre effect
that when the low bits are simply a sequential revision number (a common
though not universal pattern), then odd specified revisions must be matched
exactly, whereas even specified revisions will also match the next odd
revision, likewise for powers of 4, 8 and so forth.
To correctly do inexact matching we'd need to re-organize the table of CPU
specs to include a mask showing what PVR range the spec is compatible with
(similar to the cputable code in the Linux kernel).
For now, just remove the bogosity by only permitting exact PVR matches.
That at least makes the matching simple and consistent. If we need inexact
matching we can add the necessary per-subfamily masks later.
David Gibson [Wed, 12 Oct 2011 19:10:30 +0000 (19:10 +0000)]
pseries: Update SLOF firmware image
This patch is a general update to the SLOF firmware image used on the
pseries machine. This doesn't contain updates for specific features but
contains a number of bugfixes and enhancements in the main SLOF tree from
Thomas Huth.
David Gibson [Mon, 10 Oct 2011 18:31:01 +0000 (18:31 +0000)]
pseries: Add device tree properties for VMX/VSX and DFP under kvm
Sufficiently recent PAPR specifications define properties "ibm,vmx"
and "ibm,dfp" on the CPU node which advertise whether the VMX vector
extensions (or the later VSX version) and/or the Decimal Floating
Point operations from IBM's recent POWER CPUs are available.
Currently we do not put these in the guest device tree and the guest
kernel will consequently assume they are not available. This is good,
because they are not supported under TCG. VMX is similar enough to
Altivec that it might be trivial to support, but VSX and DFP would
both require significant work to support in TCG.
However, when running under kvm on a host which supports these
instructions, there's no reason not to let the guest use them. This
patch, therefore, checks for the relevant support on the host CPU
and, if present, advertises them to the guest as well.
David Gibson [Mon, 10 Oct 2011 18:31:00 +0000 (18:31 +0000)]
ppc: Generalize the kvmppc_get_clockfreq() function
Currently the kvmppc_get_clockfreq() function reads the host's clock
frequency from /proc/device-tree, which is useful to past to the guest
in KVM setups. However, there are some other host properties
advertised in the device tree which can also be relevant to the
guests.
This patch, therefore, replaces kvmppc_get_clockfreq() which can
retrieve any named, single integer property from the host device
tree's CPU node.
Set an invalid-bits mask for each SPE instructions
SPE instructions are defined by pairs. Currently, the invalid-bits mask is set
for the first instruction, but the second one can have a different mask.
David Gibson [Thu, 29 Sep 2011 21:39:12 +0000 (21:39 +0000)]
pseries: Use Book3S-HV TCE acceleration capabilities
The pseries machine of qemu implements the TCE mechanism used as a
virtual IOMMU for the PAPR defined virtual IO devices. Because the
PAPR spec only defines a small DMA address space, the guest VIO
drivers need to update TCE mappings very frequently - the virtual
network device is particularly bad. This means many slow exits to
qemu to emulate the H_PUT_TCE hypercall.
Sufficiently recent kernels allow this to be mitigated by implementing
H_PUT_TCE in the host kernel. To make use of this, however, qemu
needs to initialize the necessary TCE tables, and map them into itself
so that the VIO device implementations can retrieve the mappings when
they access guest memory (which is treated as a virtual DMA
operation).
This patch adds the necessary calls to use the KVM TCE acceleration.
If the kernel does not support acceleration, or there is some other
error creating the accelerated TCE table, then it will still fall back
to full userspace TCE implementation.
David Gibson [Thu, 29 Sep 2011 21:39:11 +0000 (21:39 +0000)]
pseries: Allow KVM Book3S-HV on PPC970 CPUS
At present, using the hypervisor aware Book3S-HV KVM will only work
with qemu on POWER7 CPUs. PPC970 CPUs also have hypervisor
capability, but they lack the VRMA feature which makes assigning guest
memory easier.
In order to allow KVM Book3S-HV on PPC970, we need to specially
allocate the first chunk of guest memory (the "Real Mode Area" or
RMA), so that it is physically contiguous.
Sufficiently recent host kernels allow such contiguous RMAs to be
allocated, with a kvm capability advertising whether the feature is
available and/or necessary on this hardware. This patch enables qemu
to use this support, thus allowing kvm acceleration of pseries qemu
machines on PPC970 hardware.
David Gibson [Thu, 29 Sep 2011 21:39:10 +0000 (21:39 +0000)]
pseries: Support SMT systems for KVM Book3S-HV
Alex Graf has already made qemu support KVM for the pseries machine
when using the Book3S-PR KVM variant (which runs the guest in
usermode, emulating supervisor operations). This code allows gets us
very close to also working with KVM Book3S-HV (using the hypervisor
capabilities of recent POWER CPUs).
This patch moves us another step towards Book3S-HV support by
correctly handling SMT (multithreaded) POWER CPUs. There are two
parts to this:
* Querying KVM to check SMT capability, and if present, adjusting the
cpu numbers that qemu assigns to cause KVM to assign guest threads
to cores in the right way (this isn't automatic, because the POWER
HV support has a limitation that different threads on a single core
cannot be in different guests at the same time).
* Correctly informing the guest OS of the SMT thread to core mappings
via the device tree.
this patch fix multiple issues with VirtFS tracing.
a) Add tracepoint to the correct code path. We handle error in complete_pdu
b) Fix indentation in python script
c) Fix variable naming issue in python script
Paolo Bonzini [Tue, 25 Oct 2011 10:53:37 +0000 (12:53 +0200)]
scsi-disk: add scsi-block for device passthrough
scsi-block is a new device that supports device passthrough of Linux
block devices (i.e. /dev/sda, not /dev/sg0). It uses SG_IO for commands
other than I/O commands, and regular AIO read/writes for I/O commands.
Besides being simpler to configure (no mapping required to scsi-generic
device names), this removes the need for a large bounce buffer and,
in the future, will get scatter/gather support for free from scsi-disk.
Paolo Bonzini [Tue, 25 Oct 2011 10:53:36 +0000 (12:53 +0200)]
scsi: push request restart to SCSIDevice
The request restart mechanism is generic and could be reused for
scsi-generic. In the meanwhile, pushing it to SCSIDevice avoids
that scsi_dma_restart_bh looks at SCSIGenericReqs when working on
a scsi-block device.
The code is the same that is already in hw/scsi-disk.c, with
the type flags replaced by req->cmd.mode and a more generic way to
requeue SCSI_XFER_NONE commands.
I also added a missing call to qemu_del_vm_change_state_handler.
Paolo Bonzini [Tue, 25 Oct 2011 10:53:34 +0000 (12:53 +0200)]
scsi-disk: bump SCSIRequest reference count until aio completion runs
In some cases a request may be canceled before the completion callback
runs. Keep a reference to the request between starting an AIO operation
and the corresponding scsi_req_cancel or scsi_*_complete.
When a request has to be retried, the request can be dropped because
scsi_dma_restart_bh only looks at requests that are enqueued. As such,
they always have at least a reference.
Paolo Bonzini [Tue, 25 Oct 2011 10:53:33 +0000 (12:53 +0200)]
scsi: do not call transfer_data after canceling a request
Otherwise, if cancellation is "faked" by the AIO layer and goes
through qemu_aio_flush, the whole request is completed synchronously
during scsi_req_cancel.
Using the enqueued flag would work here, but not in the next patches,
so I'm introducing a new io_canceled flag. That's because scsi_req_data
is a synchronous callback and the enqueued flag might be reset by the
time it returns. scsi-disk cannot unref the request until after calling
scsi_req_data.
projects / qemu.git / log