The ongoing QEMU multi-queue block layer effort makes it possible for multiple
threads to process I/O in parallel. The nbd block driver is not compatible with
the multi-queue block layer yet because QIOChannel cannot be used easily from
coroutines running in multiple threads. This series changes the QIOChannel API
to make that possible.
In the current API, calling qio_channel_attach_aio_context() sets the
AioContext where qio_channel_yield() installs an fd handler prior to yielding:
qio_channel_attach_aio_context(ioc, my_ctx);
...
qio_channel_yield(ioc); // my_ctx is used here
...
qio_channel_detach_aio_context(ioc);
This API design has limitations: reading and writing must be done in the same
AioContext and moving between AioContexts involves a cumbersome sequence of API
calls that is not suitable for doing on a per-request basis.
There is no fundamental reason why a QIOChannel needs to run within the
same AioContext every time qio_channel_yield() is called. QIOChannel
only uses the AioContext while inside qio_channel_yield(). The rest of
the time, QIOChannel is independent of any AioContext.
In the new API, qio_channel_yield() queries the AioContext from the current
coroutine using qemu_coroutine_get_aio_context(). There is no need to
explicitly attach/detach AioContexts anymore and
qio_channel_attach_aio_context() and qio_channel_detach_aio_context() are gone.
One coroutine can read from the QIOChannel while another coroutine writes from
a different AioContext.
This API change allows the nbd block driver to use QIOChannel from any thread.
It's important to keep in mind that the block driver already synchronizes
QIOChannel access and ensures that two coroutines never read simultaneously or
write simultaneously.
This patch updates all users of qio_channel_attach_aio_context() to the
new API. Most conversions are simple, but vhost-user-server requires a
new qemu_coroutine_yield() call to quiesce the vu_client_trip()
coroutine when not attached to any AioContext.
While the API is has become simpler, there is one wart: QIOChannel has a
special case for the iohandler AioContext (used for handlers that must not run
in nested event loops). I didn't find an elegant way preserve that behavior, so
I added a new API called qio_channel_set_follow_coroutine_ctx(ioc, true|false)
for opting in to the new AioContext model. By default QIOChannel uses the
iohandler AioHandler. Code that formerly called
qio_channel_attach_aio_context() now calls
qio_channel_set_follow_coroutine_ctx(ioc, true) once after the QIOChannel is
created.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20230830224802.493686-5-stefanha@redhat.com>
[eblake: also fix migration/rdma.c]
Signed-off-by: Eric Blake <eblake@redhat.com>
Commit 1b0578f5c4 ("qemu-pr-helper: Fix build on CentOS 7") added
code to probe for 'old' libmultipath API on CentOS 7. However since
merge commit 8c345b3e6a (June 2021) we don't build/test CentOS 7 as
it felt out of our list of supported distrib versions. Therefore we
can safely remove the 'old' API check (mostly reverting commit
1b0578f5c4, except the code got converted to meson in commit
6ec0e15d95 "meson: move libmpathpersist test").
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20230605174146.87440-2-philmd@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
do_sgio can suspend via the coroutine function thread_pool_submit_co, so it
has to be coroutine_fn as well---and the same is true of all its direct and
indirect callers.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20230309084456.304669-7-pbonzini@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
thread_pool_submit_aio() is always called on a pool taken from
qemu_get_current_aio_context(), and that is the only intended
use: each pool runs only in the same thread that is submitting
work to it, it can't run anywhere else.
Therefore simplify the thread_pool_submit* API and remove the
ThreadPool function parameter.
Signed-off-by: Emanuele Giuseppe Esposito <eesposit@redhat.com>
Message-Id: <20230203131731.851116-5-eesposit@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
MSG_PEEK peeks at the channel, The data is treated as unread and
the next read shall still return this data. This support is
currently added only for socket class. Extra parameter 'flags'
is added to io_readv calls to pass extra read flags like MSG_PEEK.
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Suggested-by: Daniel P. Berrange <berrange@redhat.com>
Signed-off-by: manish.mishra <manish.mishra@nutanix.com>
Signed-off-by: Juan Quintela <quintela@redhat.com>
Simplify the function to only return the directory path. Callers are
adjusted to use the GLib function to build paths, g_build_filename().
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20220420132624.2439741-39-marcandre.lureau@redhat.com>
Do not force exit within qemu_set_log; return bool and pass
an Error value back up the stack as per usual.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20220417183019.755276-5-richard.henderson@linaro.org>
We did this with scripts/coccinelle/use-error_fatal.cocci before, in
commit 50beeb6809 and 007b06578a. This commit cleans up rarer
variations that don't seem worth matching with Coccinelle.
Cc: Thomas Huth <thuth@redhat.com>
Cc: Cornelia Huck <cornelia.huck@de.ibm.com>
Cc: Peter Xu <peterx@redhat.com>
Cc: Juan Quintela <quintela@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Cc: Marc-André Lureau <marcandre.lureau@redhat.com>
Cc: Gerd Hoffmann <kraxel@redhat.com>
Cc: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20210720125408.387910-2-armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Currently sg_io_sense_from_errno() converts the two input parameters
'errno' and 'io_hdr' into sense code and SCSI status. Having
split the function off into scsi_sense_from_errno() and
scsi_sense_from_host_status(), both of which are available generically,
we now inline the logic in the callers so that scsi-disk and
scsi-generic will be able to pass host_status to the HBA.
Signed-off-by: Hannes Reinecke <hare@suse.de>
Message-Id: <20201116184041.60465-7-hare@suse.de>
[Put together from "scsi-disk: Add sg_io callback to evaluate status"
and what remains of "scsi: split sg_io_sense_from_errno() in two functions",
with many other fixes. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It is not needed, all the callers are just saving what was
retrieved from -trace and trace_init_file can retrieve it
on its own.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20201102115841.4017692-1-pbonzini@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
clang's C11 atomic_fetch_*() functions only take a C11 atomic type
pointer argument. QEMU uses direct types (int, etc) and this causes a
compiler error when a QEMU code calls these functions in a source file
that also included <stdatomic.h> via a system header file:
$ CC=clang CXX=clang++ ./configure ... && make
../util/async.c:79:17: error: address argument to atomic operation must be a pointer to _Atomic type ('unsigned int *' invalid)
Avoid using atomic_*() names in QEMU's atomic.h since that namespace is
used by <stdatomic.h>. Prefix QEMU's APIs with 'q' so that atomic.h
and <stdatomic.h> can co-exist. I checked /usr/include on my machine and
searched GitHub for existing "qatomic_" users but there seem to be none.
This patch was generated using:
$ git grep -h -o '\<atomic\(64\)\?_[a-z0-9_]\+' include/qemu/atomic.h | \
sort -u >/tmp/changed_identifiers
$ for identifier in $(</tmp/changed_identifiers); do
sed -i "s%\<$identifier\>%q$identifier%g" \
$(git grep -I -l "\<$identifier\>")
done
I manually fixed line-wrap issues and misaligned rST tables.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200923105646.47864-1-stefanha@redhat.com>
Replace
error_report("...: %s", ..., error_get_pretty(err));
by
error_reportf_err(err, "...: ", ...);
One of the replaced messages lacked a colon. Add it.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200505101908.6207-6-armbru@redhat.com>
Compile error reported by gcc 10.0.1:
scsi/qemu-pr-helper.c: In function ‘multipath_pr_out’:
scsi/qemu-pr-helper.c:523:32: error: array subscript <unknown> is outside array bounds of ‘struct transportid *[0]’ [-Werror=array-bounds]
523 | paramp.trnptid_list[paramp.num_transportid++] = id;
| ~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~
In file included from scsi/qemu-pr-helper.c:36:
/usr/include/mpath_persist.h:168:22: note: while referencing ‘trnptid_list’
168 | struct transportid *trnptid_list[];
| ^~~~~~~~~~~~
scsi/qemu-pr-helper.c:424:35: note: defined here ‘paramp’
424 | struct prout_param_descriptor paramp;
| ^~~~~~
This highlights an actual implementation issue in function multipath_pr_out.
The variable paramp is declared with type `struct prout_param_descriptor`,
which is a struct terminated by an empty array in mpath_persist.h:
struct transportid *trnptid_list[];
That empty array was filled with code that looked like that:
trnptid_list[paramp.descr.num_transportid++] = id;
This is an actual out-of-bounds access.
The fix is to malloc `paramp`.
Signed-off-by: Christophe de Dinechin <dinechin@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since we are actually testing for the newer capng library, rename the
symbol to match.
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The 'r' variable was accidently shadowed, and because of this
we were always passing 0 to mpath_generic_sense, instead of original
return value, which triggers an abort()
This is an attempt to fix the
https://bugzilla.redhat.com/show_bug.cgi?id=1720047
although there might be other places in the code
that trigger qemu-pr-helper crash, and this fix might
not be the root cause.
The crash was reproduced by creating an iscsi target on a test machine,
and passing it twice to the guest like that:
-blockdev node-name=idisk0,driver=iscsi,transport=...,target=...
-device scsi-block,drive=idisk0,bus=scsi0.0,bootindex=-1,scsi-id=1,lun=0,share-rw=on
-device scsi-block,drive=idisk0,bus=scsi0.0,bootindex=-1,scsi-id=1,lun=1,share-rw=on
Then in the guest, both /dev/sda and /dev/sdb were aggregated by multipath to /dev/mpatha,
which was passed to a nested guest like that
-object pr-manager-helper,id=qemu_pr_helper,path=/root/work/vm/testvm/.run/pr_helper.socket
-blockdev node-name=test,driver=host_device,filename=/dev/mapper/mpatha,pr-manager=qemu_pr_helper
-device scsi-block,drive=test,bus=scsi0.0,bootindex=-1,scsi-id=0,lun=0
The nested guest run:
sg_persist --no-inquiry -v --out --register --param-sark 0x1234 /dev/sda
Strictly speaking this is wrong configuration since qemu is where
the multipath was split, and thus the iscsi target was not aware of
multipath, and thus when libmpathpersist code rightfully tried to register
the PR key on all paths, it failed to do so.
However qemu-pr-helper should not crash in this case.
Signed-off-by: Maxim Levitsky <mlevitsk@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
No header includes qemu-common.h after this commit, as prescribed by
qemu-common.h's file comment.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190523143508.25387-5-armbru@redhat.com>
[Rebased with conflicts resolved automatically, except for
include/hw/arm/xlnx-zynqmp.h hw/arm/nrf51_soc.c hw/arm/msf2-soc.c
block/qcow2-refcount.c block/qcow2-cluster.c block/qcow2-cache.c
target/arm/cpu.h target/lm32/cpu.h target/m68k/cpu.h target/mips/cpu.h
target/moxie/cpu.h target/nios2/cpu.h target/openrisc/cpu.h
target/riscv/cpu.h target/tilegx/cpu.h target/tricore/cpu.h
target/unicore32/cpu.h target/xtensa/cpu.h; bsd-user/main.c and
net/tap-bsd.c fixed up]
This commit adds a error_init() helper which calls
g_log_set_default_handler() so that glib logs (g_log, g_warning, ...)
are handled similarly to other QEMU logs. This means they will get a
timestamp if timestamps are enabled, and they will go through the
HMP monitor if one is configured.
This commit also adds a call to error_init() to the binaries
installed by QEMU. Since error_init() also calls error_set_progname(),
this means that *-linux-user, *-bsd-user and qemu-pr-helper messages
output with error_report, info_report, ... will slightly change: they
will be prefixed by the binary name.
glib debug messages are enabled through G_MESSAGES_DEBUG similarly to
the glib default log handler.
At the moment, this change will mostly impact SPICE logging if your
spice version is >= 0.14.1. With older spice versions, this is not going
to work as expected, but will not have any ill effect, so this call is
not conditional on the SPICE version.
Signed-off-by: Christophe Fergeau <cfergeau@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20190131164614.19209-3-cfergeau@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Some versions of Clang prior to 6.0 (and some builds of clang after,
such as 6.0.1-2.fc28) fail to recognize { 0 } as a valid initializer
for a struct with subobjects when -Wmissing-braces is enabled.
https://bugs.llvm.org/show_bug.cgi?id=21689 and
https://reviews.llvm.org/rL314499 suggests this should be fixed in 6.0,
but it might not be the case for older versions or downstream versions.
For now, follow the precedent of ebf2a499 and replace the standard { 0 }
with the accepted { } to silence this warning and allow the build to
work under clang 6.0.1-2.fc28, and builds prior to 6.0.
Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Message-id: 20181127184929.20065-1-jsnow@redhat.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
After commit b3f1c8c413 "qemu-pr-helper: use new
libmultipath API", QEMU started using new libmultipath API, which is not
available on CentOS 7.x.
This fixes that by probing the new libmultipath API in configure. If it fails,
then try probing the old API. If it fails, then consider libmultipath not
available.
With this, configure script defines CONFIG_MPATH_NEW_API that is used in
scsi/qemu-pr-helper.c to use the new libmultipath API.
Fixes: b3f1c8c413
BugLink: https://bugs.launchpad.net/qemu/+bug/1786343
Signed-off-by: Murilo Opsfelder Araujo <muriloo@linux.ibm.com>
Message-Id: <20180810141116.24016-1-muriloo@linux.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
When reviewing Paolo's pr-helper patches I've noticed couple of
problems:
1) socket_path needs to be calculated at two different places
(one for printing out help, the other if socket activation is NOT
used),
2) even though the default socket_path is allocated in
compute_default_paths() it is the only default path the function
handles. For instance, pidfile is allocated outside of this
function. And yet again, at different places than 1)
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Message-Id: <c791ba035f26ea957e8f3602e3009b621769b1ba.1530611283.git.mprivozn@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
After reading a PR IN command with zero request size in prh_read_request,
the resp->result field will be uninitialized and the resp.sz field will
be also uninitialized when returning to prh_co_entry.
If resp->result == GOOD (from a previous successful reply or just luck),
then the assert in prh_write_response might not be triggered and
uninitialized response will be sent.
The fix is to remove the whole handling of sz == 0 in prh_co_entry.
Those errors apply only to PR OUT commands and it's perfectly okay to
catch them later in do_pr_out and multipath_pr_out; the check for
too-short parameters in fact doesn't apply in the easy SG_IO case, as
it can be left to the target firmware even.
The result is that prh_read_request does not fail requests anymore and
prh_co_entry becomes simpler.
Reported-by: Dima Stepanov <dimastep@yandex-team.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
The response size is expected to be zero if the SCSI status is not
"GOOD", but nothing was resetting it.
This can be reproduced simply by "sg_persist -s /dev/sdb" where /dev/sdb
in the guest is a scsi-block device corresponding to a multipath device
on the host.
Before:
PR in (Read full status): Aborted command
and on the host:
prh_write_response: Assertion `resp->sz == 0' failed.
After:
PR in (Read full status): bad field in cdb or parameter list
(perhaps unsupported service action)
Reported-by: Jiri Belka <jbelka@redhat.com>
Reviewed-by: Michal Privoznik <mprivozn@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Currently --help shows "(default '(null)')" for the -k/--socket-path
option. Fix it by getting the default path in /var/run.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Let's write pidfile even if user did not request --daemon but
they requested just --pidfile. Libvirt will use exactly this.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
After we've dropped privileges it might be not possible to write
pidfile. For instance, if this binary is run as root (because
user wants it to write pidfile to some privileged location)
writing pidfile fails because privileges are dropped before we
even get to that.
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Due to wrong specification of arguments to getopt_long() any
attempt to set pidfile resulted in:
1) the default to be leaked
2) the @pidfile variable to be set to NULL (because optarg is
NULL without this patch).
Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Message-Id: <6f10cd53d361a395aa0e85a9311ec4e9a8fc11e5.1521868451.git.mprivozn@redhat.com>
Cc: qemu-stable@nongnu.org
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Since commit 67a1de0d19 there is no space anymore between the
version number and the parentheses when running configure with
--with-pkgversion=foo :
$ qemu-system-s390x --version
QEMU emulator version 2.11.50(foo)
But the space is included when building without that option
when building from a git checkout:
$ qemu-system-s390x --version
QEMU emulator version 2.11.50 (v2.11.0-1494-gbec9c64-dirty)
The same confusion exists with the "query-version" QMP command.
Let's fix this by introducing a proper QEMU_FULL_VERSION definition
that includes the space and parentheses, while the QEMU_PKGVERSION
should just cleanly contain the package version string itself.
Note that this also changes the behavior of the "query-version" QMP
command (the space and parentheses are not included there anymore),
but that's supposed to be OK since the strings there are not meant
to be parsed by other tools.
Fixes: 67a1de0d19
Buglink: https://bugs.launchpad.net/qemu/+bug/1673373
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <1518692807-25859-1-git-send-email-thuth@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
1) Return a generic sense if TEST UNIT READY does not provide one;
2) Fix two mistakes in copying from the spec.
Cc: qemu-stable@nongnu.org
Reported-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
libmultipath has recently changed its API. The new API supports multi-threaded
clients better. Unfortunately there is no backwards-compatibility, so we just
switch to the new one. Running QEMU compiled with the new library on the old
library will likely crash, while doing the opposite will cause QEMU not to
start at all (because udev, get_multipath_config and put_multipath_config
are undefined).
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Proper support of persistent reservation for multipath devices requires
communication with the multipath daemon, so that the reservation is
registered and applied when a path comes up. The device mapper
utilities provide a library to do so; this patch makes qemu-pr-helper.c
detect multipath devices and, when one is found, delegate the operation
to libmpathpersist.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Introduce a privileged helper to run persistent reservation commands.
This lets virtual machines send persistent reservations without using
CAP_SYS_RAWIO or out-of-tree patches. The helper uses Unix permissions
and SCM_RIGHTS to restrict access to processes that can access its socket
and prove that they have an open file descriptor for a raw SCSI device.
The next patch will also correct the usage of persistent reservations
with multipath devices.
It would also be possible to support for Linux's IOC_PR_* ioctls in
the future, to support NVMe devices. For now, however, only SCSI is
supported.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
