Peter Xu [Mon, 26 Jun 2017 10:28:55 +0000 (18:28 +0800)]
migration: add "return-path" capability
When this capability is enabled, QEMU will use the return path even for
precopy migration. This is helpful at least in one case when destination
failed to load the image while source quited without confirmation. With
return path, source will wait for the last response from destination,
and if destination fails, it'll fail the migration on source, then the
guest can be run again on the source (rather than assuming to be good,
then the guest will be lost after source quits).
It needs to be enabled explicitly on source, otherwise disabled.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498472935-14461-1-git-send-email-peterx@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Halil Pasic [Fri, 23 Jun 2017 14:48:23 +0000 (16:48 +0200)]
vmstate: error hint for failed equal checks
In some cases a failing VMSTATE_*_EQUAL does not mean we detected a bug,
but it's actually the best we can do. Especially in these cases a verbose
error message is required.
Let's introduce infrastructure for specifying a error hint to be used if
equal check fails. Let's do this by adding a parameter to the _EQUAL
macros called _err_hint. Also change all current users to pass NULL as
last parameter so nothing changes for them.
Signed-off-by: Halil Pasic <pasic@linux.vnet.ibm.com>
Message-Id: <20170623144823.42936-1-pasic@linux.vnet.ibm.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Wed, 28 Jun 2017 07:15:44 +0000 (15:15 +0800)]
migration: add comment for TYPE_MIGRATE
It'll be strange that the migration object inherits TYPE_DEVICE. Add
some explanations to it.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498634144-26508-1-git-send-email-peterx@redhat.com> Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:19 +0000 (12:10 +0800)]
migration: hmp: dump globals
Now we have some globals that can be configured for migration. Dump them
in HMP info migration for better debugging.
(we can also use this to monitor whether COMPAT fields are applied
correctly on compatible machines)
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-11-git-send-email-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
are playing similar role here. This patch merges the first one into
second, then we'll have a single place to reference whether we need to
send the configuration section.
I didn't remove the MachineState.enforce_config_section field since when
applying that machine property (in machine_set_property()) we haven't
yet initialized global properties and migration object. Then, it's
still not easy to pass that boolean to MigrationState at such an early
time.
A natural benefit for current patch is that now we kept the meaning of
"enforce-config-section" since it'll still have the highest
priority (that's what "enforce" mean I guess).
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-10-git-send-email-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:17 +0000 (12:10 +0800)]
migration: move skip_section_footers
Move it into MigrationState, revert its meaning and renaming it to
send_section_footer, with a property bound to it. Same trick is played
like previous patches.
Removing savevm_skip_section_footers().
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-9-git-send-email-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:16 +0000 (12:10 +0800)]
migration: move skip_configuration out
It was in SaveState but now moved to MigrationState altogether, reverted
its meaning, then renamed to "send_configuration". Again, using
HW_COMPAT_2_3 for old PC/SPAPR machines, and accel_register_prop() for
xen_init().
Removing savevm_skip_configuration().
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-8-git-send-email-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:15 +0000 (12:10 +0800)]
migration: move only_migratable to MigrationState
One less global variable, and it does only matter with migration.
We keep the old "--only-migratable" option, but also now we support:
-global migration.only-migratable=true
Currently still keep the old interface.
Hmm, now vl.c has no way to access migrate_get_current(). Export a
function for it to setup only_migratable.
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-7-git-send-email-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:14 +0000 (12:10 +0800)]
migration: move global_state.optional out
Put it into MigrationState then we can use the properties to specify
whether to enable storing global state.
Removing global_state_set_optional() since now we can use HW_COMPAT_2_3
for x86/power, and AccelClass.global_props for Xen.
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-6-git-send-email-peterx@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:13 +0000 (12:10 +0800)]
migration: let MigrationState be a qdev
Let the old man "MigrationState" join the object family. Direct benefit
is that we can start to use all the property features derived from
current QDev, like: HW_COMPAT_* bits, command line setup for migration
parameters (so will never need to set them up each time using HMP/QMP,
this is really, really attractive for test writters), etc.
I see no reason to disallow this happen yet. So let's start from this
one, to see whether it would be anything good.
Now we init the MigrationState struct statically in main() to make sure
it's initialized after global properties are applied, since we'll use
them during creation of the object.
No functional change at all.
Reviewed-by: Juan Quintela <quintela@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-5-git-send-email-peterx@redhat.com> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:12 +0000 (12:10 +0800)]
vl: clean up global property registration
It's not that clear on how the global properties are registered to
global_props (and also its priority relationship). Let's provide a
single function to be called in main() for that, with comment to explain
it a bit.
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-4-git-send-email-peterx@redhat.com> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:11 +0000 (12:10 +0800)]
accel: introduce AccelClass.global_props
Introduce this new field for the accelerator classes so that each
specific accelerator in the future can register its own global
properties to be used further by the system. It works just like how the
old machine compatible properties do, but only tailored for
accelerators.
Introduce register_compat_props_array() for it. Export it so that it may
be used in other codes as well in the future.
Suggested-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-3-git-send-email-peterx@redhat.com> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Xu [Tue, 27 Jun 2017 04:10:10 +0000 (12:10 +0800)]
machine: export register_compat_prop()
We have HW_COMPAT_*, however that's only bound to machines, not other
things (like accelerators). Behind it, it was register_compat_prop()
that played the trick. Let's export the function for further use
outside HW_COMPAT_* magic.
Meanwhile, move it to qdev-properties.c where seems more proper (since
it'll be used not only in machine codes).
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-Id: <1498536619-14548-2-git-send-email-peterx@redhat.com> Reviewed-by: Eduardo Habkost <ehabkost@redhat.com> Signed-off-by: Juan Quintela <quintela@redhat.com>
Peter Maydell [Tue, 27 Jun 2017 15:56:55 +0000 (16:56 +0100)]
Merge remote-tracking branch 'remotes/edgar/tags/edgar/mmio-exec-v2.for-upstream' into staging
edgar/mmio-exec-v2.for-upstream
# gpg: Signature made Tue 27 Jun 2017 16:22:30 BST
# gpg: using RSA key 0x29C596780F6BCA83
# gpg: Good signature from "Edgar E. Iglesias (Xilinx key) <edgar.iglesias@xilinx.com>"
# gpg: aka "Edgar E. Iglesias <edgar.iglesias@gmail.com>"
# Primary key fingerprint: AC44 FEDC 14F7 F1EB EDBF 4151 29C5 9678 0F6B CA83
* remotes/edgar/tags/edgar/mmio-exec-v2.for-upstream:
xilinx_spips: allow mmio execution
exec: allow to get a pointer for some mmio memory region
introduce mmio_interface
qdev: add MemoryRegion property
cputlb: fix the way get_page_addr_code fills the tlb
cputlb: move get_page_addr_code
cputlb: cleanup get_page_addr_code to use VICTIM_TLB_HIT
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
KONRAD Frederic [Thu, 20 Oct 2016 09:09:53 +0000 (11:09 +0200)]
xilinx_spips: allow mmio execution
This allows to execute from the lqspi area.
When the request_ptr is called the device loads 1024bytes from the SPI device.
Then this code can be executed by the guest.
Tested-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
KONRAD Frederic [Wed, 19 Oct 2016 13:06:49 +0000 (15:06 +0200)]
exec: allow to get a pointer for some mmio memory region
This introduces a special callback which allows to run code from some MMIO
devices.
SysBusDevice with a MemoryRegion which implements the request_ptr callback will
be notified when the guest try to execute code from their offset. Then it will
be able to eg: pre-load some code from an SPI device or ask a pointer from an
external simulator, etc..
When the pointer or the data in it are no longer valid the device has to
invalidate it.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
KONRAD Frederic [Thu, 16 Feb 2017 09:27:00 +0000 (10:27 +0100)]
introduce mmio_interface
This introduces mmio_interface object which contains a MemoryRegion
and can be hotplugged/hotunplugged.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
KONRAD Frederic [Thu, 16 Feb 2017 14:06:24 +0000 (15:06 +0100)]
qdev: add MemoryRegion property
We need to pass a pointer to a MemoryRegion for mmio_interface.
So this just adds that.
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
KONRAD Frederic [Fri, 3 Feb 2017 15:32:12 +0000 (16:32 +0100)]
cputlb: fix the way get_page_addr_code fills the tlb
get_page_addr_code(..) does a cpu_ldub_code to fill the tlb:
This can lead to some side effects if a device is mapped at this address.
So this patch replaces the cpu_memory_ld by a tlb_fill.
Reviewed-by: Richard Henderson <rth@twiddle.net> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
KONRAD Frederic [Fri, 3 Feb 2017 15:29:50 +0000 (16:29 +0100)]
cputlb: move get_page_addr_code
This just moves the code before VICTIM_TLB_HIT macro definition
so we can use it.
Reviewed-by: Richard Henderson <rth@twiddle.net> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
KONRAD Frederic [Fri, 3 Feb 2017 15:27:49 +0000 (16:27 +0100)]
cputlb: cleanup get_page_addr_code to use VICTIM_TLB_HIT
This replaces env1 and page_index variables by env and index
so we can use VICTIM_TLB_HIT macro later.
Reviewed-by: Richard Henderson <rth@twiddle.net> Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com> Signed-off-by: KONRAD Frederic <fred.konrad@greensocs.com> Signed-off-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Peter Maydell [Mon, 26 Jun 2017 14:38:29 +0000 (15:38 +0100)]
Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging
Block layer patches
# gpg: Signature made Mon 26 Jun 2017 14:07:32 BST
# gpg: using RSA key 0x7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>"
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74 56FE 7F09 B272 C88F 2FD6
* remotes/kevin/tags/for-upstream: (60 commits)
qemu-img: don't shadow opts variable in img_dd()
block: Do not strcmp() with NULL uri->scheme
blkverify: Catch bs->exact_filename overflow
blkdebug: Catch bs->exact_filename overflow
fix: avoid an infinite loop or a dangling pointer problem in img_commit
block: change variable names in BlockDriverState
block: Remove bdrv_aio_readv/writev/flush()
qed: Use bdrv_co_* for coroutine_fns
qed: Add coroutine_fn to I/O path functions
qed: Use a coroutine for need_check_timer
qed: Simplify request handling
qed: Use CoQueue for serialising allocations
qed: Implement .bdrv_co_readv/writev
qed: Remove recursion in qed_aio_next_io()
qed: Remove ret argument from qed_aio_next_io()
qed: Add return value to qed_aio_read/write_data()
qed: Add return value to qed_aio_write_inplace/alloc()
qed: Add return value to qed_aio_write_cow()
qed: Add return value to qed_aio_write_main()
qed: Add return value to qed_aio_write_l2_update()
...
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Kevin Wolf [Mon, 26 Jun 2017 12:57:27 +0000 (14:57 +0200)]
Merge remote-tracking branch 'mreitz/tags/pull-block-2017-06-26' into queue-block
Block patches for the block queue
# gpg: Signature made Mon Jun 26 14:56:24 2017 CEST
# gpg: using RSA key 0xF407DB0061D5CF40
# gpg: Good signature from "Max Reitz <mreitz@redhat.com>"
# Primary key fingerprint: 91BE B60A 30DB 3E88 57D1 1829 F407 DB00 61D5 CF40
* mreitz/tags/pull-block-2017-06-26:
qemu-img: don't shadow opts variable in img_dd()
block: Do not strcmp() with NULL uri->scheme
blkverify: Catch bs->exact_filename overflow
blkdebug: Catch bs->exact_filename overflow
fix: avoid an infinite loop or a dangling pointer problem in img_commit
block: change variable names in BlockDriverState
Stefan Hajnoczi [Mon, 19 Jun 2017 15:00:02 +0000 (16:00 +0100)]
qemu-img: don't shadow opts variable in img_dd()
It's confusing when two different variables have the same name in one
function.
Cc: Reda Sallahi <fullmanet@gmail.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-id: 20170619150002.3033-1-stefanha@redhat.com Signed-off-by: Max Reitz <mreitz@redhat.com>
Max Reitz [Tue, 13 Jun 2017 20:57:26 +0000 (22:57 +0200)]
block: Do not strcmp() with NULL uri->scheme
uri_parse(...)->scheme may be NULL. In fact, probably every field may be
NULL, and the callers do test this for all of the other fields but not
for scheme (except for block/gluster.c; block/vxhs.c does not access
that field at all).
We can easily fix this by using g_strcmp0() instead of strcmp().
Cc: qemu-stable@nongnu.org Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170613205726.13544-1-mreitz@redhat.com Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
Max Reitz [Tue, 13 Jun 2017 17:20:06 +0000 (19:20 +0200)]
blkverify: Catch bs->exact_filename overflow
The bs->exact_filename field may not be sufficient to store the full
blkverify node filename. In this case, we should not generate a filename
at all instead of an unusable one.
Cc: qemu-stable@nongnu.org Reported-by: Qu Wenruo <quwenruo@cn.fujitsu.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170613172006.19685-3-mreitz@redhat.com Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
Max Reitz [Tue, 13 Jun 2017 17:20:05 +0000 (19:20 +0200)]
blkdebug: Catch bs->exact_filename overflow
The bs->exact_filename field may not be sufficient to store the full
blkdebug node filename. In this case, we should not generate a filename
at all instead of an unusable one.
Cc: qemu-stable@nongnu.org Reported-by: Qu Wenruo <quwenruo@cn.fujitsu.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20170613172006.19685-2-mreitz@redhat.com Reviewed-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
Change the 'int count' parameter in *pwrite_zeros, *pdiscard related
functions (and some others) to 'int bytes', as they both refer to bytes.
This helps with code legibility.
Signed-off-by: Manos Pitsidianakis <el13635@mail.ntua.gr>
Message-id: 20170609101808.13506-1-el13635@mail.ntua.gr Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Max Reitz <mreitz@redhat.com>
Kevin Wolf [Fri, 16 Jun 2017 12:43:19 +0000 (14:43 +0200)]
qed: Use bdrv_co_* for coroutine_fns
All functions that are marked coroutine_fn can directly call the
bdrv_co_* version of functions instead of going through the wrapper.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Manos Pitsidianakis <el13635@mail.ntua.gr> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Mon, 12 Jun 2017 09:12:41 +0000 (11:12 +0200)]
qed: Add coroutine_fn to I/O path functions
Now that we stay in coroutine context for the whole request when doing
reads or writes, we can add coroutine_fn annotations to many functions
that can do I/O or yield directly.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Fri, 18 Nov 2016 15:04:59 +0000 (16:04 +0100)]
qed: Use a coroutine for need_check_timer
This fixes the last place where we degraded from AIO to actual blocking
synchronous I/O requests. Putting it into a coroutine means that instead
of blocking, the coroutine simply yields while doing I/O.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Fri, 18 Nov 2016 13:47:36 +0000 (14:47 +0100)]
qed: Simplify request handling
Now that we process a request in the same coroutine from beginning to
end and don't drop out of it any more, we can look like a proper
coroutine-based driver and simply call qed_aio_next_io() and get a
return value from it instead of spawning an additional coroutine that
reenters the parent when it's done.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Fri, 18 Nov 2016 14:32:17 +0000 (15:32 +0100)]
qed: Use CoQueue for serialising allocations
Now that we're running in coroutine context, the ad-hoc serialisation
code (which drops a request that has to wait out of coroutine context)
can be replaced by a CoQueue.
This means that when we resume a serialised request, it is running in
coroutine context again and its I/O isn't blocking any more.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Mon, 14 Nov 2016 13:20:00 +0000 (14:20 +0100)]
qed: Implement .bdrv_co_readv/writev
Most of the qed code is now synchronous and matches the coroutine model.
One notable exception is the serialisation between requests which can
still schedule a callback. Before we can replace this with coroutine
locks, let's convert the driver's external interfaces to the coroutine
versions.
We need to be careful to handle both requests that call the completion
callback directly from the calling coroutine (i.e. fully synchronous
code) and requests that involve some callback, so that we need to yield
and wait for the completion callback coming from outside the coroutine.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Manos Pitsidianakis <el13635@mail.ntua.gr> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Thu, 17 Nov 2016 14:40:41 +0000 (15:40 +0100)]
qed: Add return value to qed_aio_write_cow()
Don't recurse into qed_aio_next_io() and qed_aio_complete() here, but
just return an error code and let the caller handle it.
While refactoring qed_aio_write_alloc() to accomodate the change,
qed_aio_write_zero_cluster() ended up with a single line, so I chose to
inline that line and remove the function completely.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Tue, 15 Nov 2016 10:14:01 +0000 (11:14 +0100)]
qed: Remove callback from qed_copy_from_backing_file()
With this change, qed_aio_write_prefill() and qed_aio_write_postfill()
collapse into a single function. This is reflected by a rename of the
combined function to qed_aio_write_cow().
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Kevin Wolf [Wed, 16 Nov 2016 16:31:14 +0000 (17:31 +0100)]
qed: Use bottom half to resume waiting requests
The qed driver serialises allocating write requests. When the active
allocation is finished, the AIO callback is called, but after this, the
next allocating request is immediately processed instead of leaving the
coroutine. Resuming another allocation request in the same request
coroutine means that the request now runs in the wrong coroutine.
The following is one of the possible effects of this: The completed
request will generally reenter its request coroutine in a bottom half,
expecting that it completes the request in bdrv_driver_pwritev().
However, if the second request actually yielded before leaving the
coroutine, the reused request coroutine is in an entirely different
place and is reentered prematurely. Not a good idea.
Let's make sure that we exit the coroutine after completing the first
request by resuming the next allocating request only with a bottom
half.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Alberto Garcia [Tue, 20 Jun 2017 13:01:36 +0000 (16:01 +0300)]
qcow2: Use offset_into_cluster() and offset_to_l2_index()
We already have functions for doing these calculations, so let's use
them instead of doing everything by hand. This makes the code a bit
more readable.
Signed-off-by: Alberto Garcia <berto@igalia.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:08 +0000 (16:40 +0300)]
qcow2: Merge the writing of the COW regions with the guest data
If the guest tries to write data that results on the allocation of a
new cluster, instead of writing the guest data first and then the data
from the COW regions, write everything together using one single I/O
operation.
This can improve the write performance by 25% or more, depending on
several factors such as the media type, the cluster size and the I/O
request size.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:07 +0000 (16:40 +0300)]
qcow2: Pass a QEMUIOVector to do_perform_cow_{read,write}()
Instead of passing a single buffer pointer to do_perform_cow_write(),
pass a QEMUIOVector. This will allow us to merge the write requests
for the COW regions and the actual data into a single one.
Although do_perform_cow_read() does not strictly need to change its
API, we're doing it here as well for consistency.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:06 +0000 (16:40 +0300)]
qcow2: Allow reading both COW regions with only one request
Reading both COW regions requires two separate requests, but it's
perfectly possible to merge them and perform only one. This generally
improves performance, particularly on rotating disk drives. The
downside is that the data in the middle region is read but discarded.
This patch takes a conservative approach and only merges reads when
the size of the middle region is <= 16KB.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:05 +0000 (16:40 +0300)]
qcow2: Split do_perform_cow() into _read(), _encrypt() and _write()
This patch splits do_perform_cow() into three separate functions to
read, encrypt and write the COW regions.
perform_cow() can now read both regions first, then encrypt them and
finally write them to disk. The memory allocation is also done in
this function now, using one single buffer large enough to hold both
regions.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:04 +0000 (16:40 +0300)]
qcow2: Make perform_cow() call do_perform_cow() twice
Instead of calling perform_cow() twice with a different COW region
each time, call it just once and make perform_cow() handle both
regions.
This patch simply moves code around. The next one will do the actual
reordering of the COW operations.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:03 +0000 (16:40 +0300)]
qcow2: Use unsigned int for both members of Qcow2COWRegion
Qcow2COWRegion has two attributes:
- The offset of the COW region from the start of the first cluster
touched by the I/O request. Since it's always going to be positive
and the maximum request size is at most INT_MAX, we can use a
regular unsigned int to store this offset.
- The size of the COW region in bytes. This is guaranteed to be >= 0,
so we should use an unsigned type instead.
In x86_64 this reduces the size of Qcow2COWRegion from 16 to 8 bytes.
It will also help keep some assertions simpler now that we know that
there are no negative numbers.
The prototype of do_perform_cow() is also updated to reflect these
changes.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Mon, 19 Jun 2017 13:40:02 +0000 (16:40 +0300)]
qcow2: Remove unused Error variable in do_perform_cow()
We are using the return value of qcow2_encrypt_sectors() to detect
problems but we are throwing away the returned Error since we have no
way to report it to the user. Therefore we can simply get rid of the
local Error variable and pass NULL instead.
Alternatively we could try to figure out a way to pass the original
error instead of simply returning -EIO, but that would be more
invasive, so let's keep the current approach.
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Stephen Bates [Tue, 13 Jun 2017 10:08:35 +0000 (04:08 -0600)]
nvme: Add support for Read Data and Write Data in CMBs.
Add the ability for the NVMe model to support both the RDS and WDS
modes in the Controller Memory Buffer.
Although not currently supported in the upstreamed Linux kernel a fork
with support exists [1] and user-space test programs that build on
this also exist [2].
Useful for testing CMB functionality in preperation for real CMB
enabled NVMe devices (coming soon).
Signed-off-by: Stephen Bates <sbates@raithlin.com> Reviewed-by: Logan Gunthorpe <logang@deltatee.com> Reviewed-by: Keith Busch <keith.busch@intel.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Stefan Hajnoczi [Thu, 15 Jun 2017 16:38:10 +0000 (17:38 +0100)]
migration: hold AioContext lock for loadvm qemu_fclose()
migration_incoming_state_destroy() uses qemu_fclose() on the vmstate
file. Make sure to call it inside an AioContext acquire/release region.
This fixes an 'qemu: qemu_mutex_unlock: Operation not permitted' abort
in loadvm.
This patch closes the vmstate file before ending the drained region.
Previously we closed the vmstate file after ending the drained region.
The order does not matter.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Alberto Garcia [Tue, 13 Jun 2017 21:16:12 +0000 (00:16 +0300)]
throttle: Update throttle-groups.c documentation
There used to be throttle_timers_{detach,attach}_aio_context() calls
in bdrv_set_aio_context(), but since 7ca7f0f6db1fedd28d490795d778cf239
they are now in blk_set_aio_context().
Signed-off-by: Alberto Garcia <berto@igalia.com> Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Kevin Wolf [Thu, 22 Sep 2016 15:24:38 +0000 (17:24 +0200)]
doc: Document driver-specific -blockdev options
This documents the driver-specific options for the raw, qcow2 and file
block drivers for the man page. For everything else, we refer to the
QAPI documentation.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
Kevin Wolf [Thu, 22 Sep 2016 14:53:24 +0000 (16:53 +0200)]
doc: Document generic -blockdev options
This adds documentation for the -blockdev options that apply to all
nodes independent of the block driver used.
All options that are shared by -blockdev and -drive are now explained in
the section for -blockdev. The documentation of -drive mentions that all
-blockdev options are accepted as well.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
Stefan Hajnoczi [Mon, 22 May 2017 13:57:04 +0000 (14:57 +0100)]
migration: use bdrv_drain_all_begin/end() instead bdrv_drain_all()
blk/bdrv_drain_all() only takes effect for a single instant and then
resumes block jobs, guest devices, and other external clients like the
NBD server. This can be handy when performing a synchronous drain
before terminating the program, for example.
Monitor commands usually need to quiesce I/O across an entire code
region so blk/bdrv_drain_all() is not suitable. They must use
bdrv_drain_all_begin/end() to mark the region. This prevents new I/O
requests from slipping in or worse - block jobs completing and modifying
the graph.
I audited other blk/bdrv_drain_all() callers but did not find anything
that needs a similar fix. This patch fixes the savevm/loadvm commands.
Although I haven't encountered a read world issue this makes the code
safer.
Suggested-by: Kevin Wolf <kwolf@redhat.com> Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Stefan Hajnoczi [Mon, 22 May 2017 13:57:03 +0000 (14:57 +0100)]
migration: avoid recursive AioContext locking in save_vmstate()
AioContext was designed to allow nested acquire/release calls. It uses
a recursive mutex so callers don't need to worry about nesting...or so
we thought.
BDRV_POLL_WHILE() is used to wait for block I/O requests. It releases
the AioContext temporarily around aio_poll(). This gives IOThreads a
chance to acquire the AioContext to process I/O completions.
It turns out that recursive locking and BDRV_POLL_WHILE() don't mix.
BDRV_POLL_WHILE() only releases the AioContext once, so the IOThread
will not be able to acquire the AioContext if it was acquired
multiple times.
Instead of trying to release AioContext n times in BDRV_POLL_WHILE(),
this patch simply avoids nested locking in save_vmstate(). It's the
simplest fix and we should step back to consider the big picture with
all the recent changes to block layer threading.
This patch is the final fix to solve 'savevm' hanging with -object
iothread.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Stefan Hajnoczi [Mon, 22 May 2017 13:57:02 +0000 (14:57 +0100)]
block: use BDRV_POLL_WHILE() in bdrv_rw_vmstate()
Calling aio_poll() directly may have been fine previously, but this is
the future, man! The difference between an aio_poll() loop and
BDRV_POLL_WHILE() is that BDRV_POLL_WHILE() releases the AioContext
around aio_poll().
This allows the IOThread to run fd handlers or BHs to complete the
request. Failure to release the AioContext causes deadlocks.
Using BDRV_POLL_WHILE() partially fixes a 'savevm' hang with -object
iothread.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Stefan Hajnoczi [Mon, 22 May 2017 13:57:01 +0000 (14:57 +0100)]
block: count bdrv_co_rw_vmstate() requests
Call bdrv_inc/dec_in_flight() for vmstate reads/writes. This seems
unnecessary at first glance because vmstate reads/writes are done
synchronously while the guest is stopped. But we need the bdrv_wakeup()
in bdrv_dec_in_flight() so the main loop sees request completion.
Besides, it's cleaner to count vmstate reads/writes like ordinary
read/write requests.
The bdrv_wakeup() partially fixes a 'savevm' hang with -object iothread.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Paolo Bonzini <pbonzini@redhat.com> Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Kevin Wolf [Fri, 9 Jun 2017 11:37:01 +0000 (13:37 +0200)]
qemu-iotests: Test exiting qemu with running job
When qemu is exited, all running jobs should be cancelled successfully.
This adds a test for this for all types of block jobs that currently
exist in qemu.
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com>
Kevin Wolf [Fri, 9 Jun 2017 11:32:48 +0000 (13:32 +0200)]
qemu-iotests: Allow starting new qemu after cleanup
After _cleanup_qemu(), test cases should be able to start the next qemu
process and call _cleanup_qemu() for that one as well. For this to work
cleanly, we need to improve the cleanup so that the second invocation
doesn't try to kill the qemu instances from the first invocation a
second time (which would result in error messages).
Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
Kevin Wolf [Fri, 9 Jun 2017 11:29:36 +0000 (13:29 +0200)]
commit: Fix completion with extra reference
commit_complete() can't assume that after its block_job_completed() the
job is actually immediately freed; someone else may still be holding
references. In this case, the op blockers on the intermediate nodes make
the graph reconfiguration in the completion code fail.
Call block_job_remove_all_bdrv() manually so that we know for sure that
any blockers on intermediate nodes are given up.
Cc: qemu-stable@nongnu.org Signed-off-by: Kevin Wolf <kwolf@redhat.com> Reviewed-by: Eric Blake <eblake@redhat.com> Reviewed-by: Max Reitz <mreitz@redhat.com>
Peter Maydell [Fri, 2 Jun 2017 14:35:38 +0000 (15:35 +0100)]
configure: Define NCURSES_WIDECHAR if we're using curses
We want the wide character functions from the ncurses header.
Unfortunately it doesn't provide them by default, but only
if either:
* NCURSES_WIDECHAR is defined (for ncurses 20111030 and up)
* _XOPEN_SOURCE/_XOPEN_SOURCE_EXTENDED are suitably defined
So far we have been implicitly relying on the latter, because
for GNU libc when we define _GNU_SOURCE this causes libc
to define the _XOPEN_SOURCE macros for us. Unfortunately
this doesn't work on all libcs, because some (like OSX and
musl libc) do not define _XOPEN_SOURCE when _GNU_SOURCE
is defined.
We can't fix this by defining _XOPEN_SOURCE ourselves, because
that also means "and don't provide any functions that aren't in
that standard", and not all libcs provide any way to override
that to also get the non-standard functions. In particular
FreeBSD has no such mechanism, and OSX's _DARWIN_C_SOURCE
doesn't reenable everything (for instance getpagesize()
is still not prototyped if _DARWIN_C_SOURCE and _XOPEN_SOURCE
are both defined).
So we have to define NCURSES_WIDECHAR. (This will only work
if your ncurses is at least 20111030, as older versions
don't honour this macro.)
Signed-off-by: Peter Maydell <peter.maydell@linaro.org> Acked-by: Laszlo Ersek <lersek@redhat.com>
Message-id: 1496414138-7622-1-git-send-email-peter.maydell@linaro.org
Peter Maydell [Fri, 23 Jun 2017 17:11:48 +0000 (18:11 +0100)]
Merge remote-tracking branch 'remotes/rth/tags/pull-s390-20170623' into staging
Queued target/s390x patches
# gpg: Signature made Fri 23 Jun 2017 17:18:24 BST
# gpg: using RSA key 0xAD1270CC4DD0279B
# gpg: Good signature from "Richard Henderson <rth7680@gmail.com>"
# gpg: aka "Richard Henderson <rth@redhat.com>"
# gpg: aka "Richard Henderson <rth@twiddle.net>"
# Primary key fingerprint: 9CB1 8DDA F8E8 49AD 2AFC 16A4 AD12 70CC 4DD0 279B
* remotes/rth/tags/pull-s390-20170623:
target/s390x: Implement idte instruction
target/s390x: Improve heuristic for ipte
target/s390x: Indicate and check for local tlb clearing
target/s390x: Clean up TB flag bits
target/s390x: Finish implementing ETF2-ENH
target/s390x: Mark STFLE_49 facility as available
target/s390x: Implement processor-assist insn
target/s390x: Implement execution-hint insns
target/s390x: Mark STFLE_53 facility as available
target/s390x: Implement load-and-zero-rightmost-byte insns
target/s390x: Implement load-on-condition-2 insns
target/s390x: Mark FPSEH facility as available
target/s390x: implement mvcos instruction
target/s390x: change PSW_SHIFT_KEY
target/s390x: Map existing FAC_* names to S390_FEAT_* names
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Let's keep it very simple for now and flush the complete tlb,
we currently can't find the right entries in our tlb, we would have
to store the used tables for each element.
As we now fully implement the DAT-enhancement facility, we can allow to
enable it for the qemu CPU model.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170622094151.28633-4-david@redhat.com> Signed-off-by: Richard Henderson <rth@twiddle.net>
If only the page index is set, most likely we don't have a valid
virtual address. Let's do a full tlb flush for that case.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170622094151.28633-3-david@redhat.com> Signed-off-by: Richard Henderson <rth@twiddle.net>
target/s390x: Indicate and check for local tlb clearing
Let's allow to enable it for the qemu cpu model and correctly emulate
it.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170622094151.28633-2-david@redhat.com> Signed-off-by: Richard Henderson <rth@twiddle.net>
Most of the PSW bits that were being copied into TB->flags
are not relevant to translation. Removing those that are
unnecessary reduces the amount of translation required.
This adds support for the MOVE WITH OPTIONAL SPECIFICATIONS (MVCOS)
instruction. Allow to enable it for the qemu cpu model using
qemu-system-s390x ... -cpu qemu,mvcos=on ...
This allows to boot linux kernel that uses it for uacccess.
We are missing (as for most other part) low address protection checks,
PSW key / storage key checks and support for AR-mode.
We fake an ADDRESSING exception when called from problem state (which
seems to rely on PSW key checks to be in place) and if AR-mode is used.
user mode will always see a PRIVILEDGED exception.
This patch is based on an original patch by Miroslav Benes (thanks!).
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20170614133819.18480-3-david@redhat.com> Signed-off-by: Richard Henderson <rth@twiddle.net>
projects / mirror_qemu.git / log