Introduce a new build variable "nosharedlibs".
In tools/libxc use it instead of $(stubdom).
In tools/xenstore honour it, and build static clients.
If shared libs are disabled, do not try to install or symlink them.
Set nosharedlibs when building for MiniOS or NetBSDRump.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
---
v2: Clarify deliberate INSTALL_SHLIB and SYMLINK_SHLIB breakage.
This variable is the suffix to use for finding libraries when doing
compile-time linking. For now we always set it to ".so" - so no
functional change.
In a forthcoming patch it may take on different values.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
INSTALL_SHLIB is like INSTALL_PROG but used only for shared libraries.
SYMLINK_SHLIB is the ln -sf rune for shared library symlinks.
Use these in the appropriate places in tools/libxc and tools/xenstore.
No functional change right now. In a forthcoming patch these
variables might take on different values.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
In principle it would be possible to make a rumpuser-xen-based stub
xenstored, but all the necessary pieces do not yet exist.
So provide a facility to disable compilation of xenstored, and use it
to disable it on rump kernels.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Do not try to support the dlopen-based xc indirection.
Introduce a local #define DO_DYNAMIC_OSDEP to centralise the condition.
Add comments to the #endifs.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
---
v2: Remove duplication by introducing DO_DYNAMIC_OSDEP.
Add comments to the #endifs.
do_ls is recursive. It had a buffer of size around 5K allocated on
the stack. This combination is not a very good idea: some
environments (eg, Mini-OS) have limited stack sizes (eg 64K).
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
In 417e6b70 I overlooked the requirement to rerun bison/flex. Do that
now. The changes are exactly those which are the result of 417e6b70.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
This updates libxlu_cfg_y.[ch] to code generated by bison from
Debian wheezy (1:2.5.dfsg-2.1 i386).
There should be no functional change since there is no change to the
source file, but we will inherit bugfixes and behavioural changes from
the new version of bison. So this is more a matter of hope than
knowledge.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
do_daemonize had open-coded handling of the results from xl_waitpid.
Instead, break out the meat of console_child_report into a new
function child_report (which returns an error code), and use it.
No functional change other than a change to the message printed if
forking the daemonic child fails.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
vncviewer_child_report was very similar to console_child_report.
We can abolish vncviewer_child_report by adding an xlchildnum
parameter to console_child_report.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
We record the descriptive string for the child in the children[]
array, and use it when reporting the exit status.
The only functional change is that the message reported for the
migration child is changed from "migration target process" to
"migration transport process".
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Compute a -v option to pass to the migration receiver.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
---
v2: Use minmsglevel_default to initialise minmsglevel.
If we ourselves are using cr-based overwriting for logging to stderr,
pass -t to the migration receiver so that it knows to do the same
(since its stderr is normally the pipe from sshd).
This requires, of course, that the receiver support that option. This
is OK from a compatibility point of view because we support migration
to newer, but not necessarily to older, versions. (If unsupported
backwards migration is still desired the use of -s "" allows the
remote invocation rune to be overridden by a command of one's choice.)
This fixes a regression introduced in 2f80ac9c0e, where migration
messages from the receiver would not use of the overwriting protocol.
CC: Olaf Hering <olaf@aepfle.de>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Provide flags
XTL_STDIOSTREAM_PROGRESS_USE_CR
XTL_STDIOSTREAM_PROGRESS_NO_CR
to allow the caller to force, or disable, the use of \r-based
overwriting of progress messages.
In the implementation, rename the variable "tty" to "progress_use_cr".
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Olaf Hering <olaf@aepfle.de>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Plumb hotplug scripts' stdout to stderr. That way if they print
anything (which really they shouldn't), it won't get mixed up with
the application's stdout. (Eg, perhaps with an xl migration
stream...)
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Roger Pau Monné <roger.pau@citrix.com>
CC: Vasiliy Tolstov <v.tolstov@selfip.ru>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Since 28d386fc43 (XSA-57), libxl writes an empty value for the
console tty node, with read-only permission for the guest, when
setting up pv console "frontends". (The actual tty value is later set
by xenconsoled.) Writing an empty node is not strictly necessary to
stop the frontend from writing dangerous values here, but it is a good
belt-and-braces approach.
Unfortunately this confuses xenconsole. It reads the empty value, and
tries to open it as the tty. xenconsole then exits.
Fix this by having xenconsole treat an empty value the same way as no
value at all.
Also, make the error opening the tty be nonfatal: we just print a
warning, but do not exit. I think this is helpful in theoretical
situations where xenconsole is racing with libxl and/or xenconsoled.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
CC: George Dunlap <george.dunlap@eu.citrix.com>
---
v2: Combine two conditions and move the free
If xenconsole (the client program) fails, it calls err. This would
previously neglect to reset the user's terminal to sanity. Use atexit
to do so.
This routinely happens in Xen 4.4 RC5 with pygrub because libxl
writes the value "" to the tty xenstore key when using xenconsole.
After this patch this just results in a harmless error message.
Reported-by: M A Young <m.a.young@durham.ac.uk>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: M A Young <m.a.young@durham.ac.uk>
CC: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v2: Fix whitespace error (reintroduce hard tab)
Fix commit message not to claim ignorance about root cause
Declaring a formal parameter to have an array type doesn't result in
the parameter actually having an array type. The type is "adjusted"
to a pointer. (C99 6.9.1(7), 6.7.5.3.)
So the use of sizeof in xc_suspend.c:get_suspend_file was wrong.
Instead, use the #define. Also get rid of the array size, as it is
misleading.
Newer versions of gcc warn about the erroneous code:
xc_suspend.c:39:25: error: argument to 'sizeof' in 'snprintf' call
is the same expression as the destination; did you mean to provide
an explicit length? [-Werror=sizeof-pointer-memaccess]
Reported-By: Julien Grall <julien.grall@linaro.org>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Ian Campbell <ian.campbell@citrix.com>
CC: Julien Grall <julien.grall@linaro.org>
--
v2: Actually change the declaration of buf.
When negotiating guest suspend via the evtchn ("fast") protocol,
the guest may still fail to respond.
So set the timeout. The existing error path will already properly
tear down our (event channel) wait.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
When negotiating guest suspend via the evtchn ("fast") protocol,
abolish synchronous wait for domain suspend.
If the guest supports the event channel suspend protocol, we used to
sit in a loop in xc_await_suspend waiting (perhaps indefinitely) for
it to suspend.
Instead, use the new libxl event channel event facility. When we see
that the event is signaled, we look at the domain to see if it has
suspended. (In this patch we do not yet set a timeout; that will come
next.)
So the suspend operation no longer blocks with the libxl ctx lock
held, and instead returns to the event loop. Additionally, domains
which signal the event channel themselves, or undergo other state
changes, will be handled more correctly.
We end up making a few more hypercalls.
Also, if we encounter errors setting up the suspend event channel
(which should not happen), abort the operation rather than falling
back to the xenstore protocol.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Improve commit message.
When we are waiting for a guest to suspend, this suspend operation
would continue to wait (until the timeout) if the guest was destroyed
or shut down for another reason, or if xc_domain_getinfolist failed.
Handle these cases correctly, as errors.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Remove unwanted error call after a new "goto err".
Replace the use of a loop with usleep().
Instead, use a xenstore watch and an event system timeout. (xenstore
fires watches on @releaseDomain when a domain shuts down.)
The logic which checks for the state of the domain is unchanged, and
not ideal, but we will leave that for the next patch.
There is not intended to be any semantic change, other than to make
the algorithm properly asynchronous and the consequential waiting be
on xenstore, rather than polling.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Remove some trailing whitespace
Improve commit message.
v3X: Do NOT use an xswait instead of separate watch and timeout.
When negotiating guest suspend via the xenstore pvcontrol protocol
(ie when the guest does NOT support the evtchn fast suspend protocol):
Replace the use of loops and usleep with a call to libxl__xswait.
Also, replace the xenstore transaction loop with one using
libxl__xs_transaction_start et al.
There is not intended to be any semantic change, other than to make
the algorithm properly asynchronous.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Add a comment to clarify last minute ack.
v3X: Do NOT rename "pvcontrol" xswait state struct to "guest_wait"
(because we're NOT going to use it for the event channel based wait
too).
In domain_suspend_callback_common, use libxl__xs_transaction_start in
a loop, rather than xs_transaction_start and a goto label.
This will improve the error handling, but have no other semantic
effect.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Factor out domain_suspend_pvcontrol_acked.
This replaces a bunch of open-coded strcmp()s and makes the code
clearer. It also eliminates the need to check for state==NULL each
time it's read, because we can check for NULL once before the strcmp.
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Improve comment re xswatch state ENOENT
Factor out the pv control node xenstore path calculation into
libxl__domain_pvcontrol_xspath.
This xs path calculation was open coded in
libxl__domain_pvcontrol_read and _write. This is undesirable because
it duplicates the code and because it makes the path inaccessible to
other parts of libxl (which are soon going to want it).
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Make domain_suspend_callback_common more callback-oriented:
* Turn the functionality behind the goto labels "err" and
"guest_suspended" into functions which can be called just before
"return".
* Deindent the "issuing %s suspend request via XenBus control node"
branch; it is going to be split up into various functions as the
xenstore work becomes callback-based.
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Make domain_suspend_callback_common do its work and then call
dss->callback_common_done, rather than simply returning its answer.
This is preparatory to abolishing the usleeps in this function and
replacing them with use of the event machinery.
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Remove some trailing whitespace
Mark the suspend callback libxl__domain_suspend_callback as
asynchronous in the helper stub generator (libxl_save_msgs_gen.pl).
We are going to want to provide an asynchronous version of this
function to get rid of the usleeps and waiting loops in the suspend
code.
libxl__domain_suspend_common_callback, the common synchronous core,
which used to be provided directly as the callback function for the
helper machinery, becomes libxl__domain_suspend_callback_common. It
can now take a typesafe parameter.
For now, provide two very similar asynchronous wrappers for it
(normal, and remus). Each is a simple function which contains only
boilerplate, calls the common synchronous core, and returns the
asynchronous response.
Essentially, we have just moved (in the case of suspend callbacks) the
call site of libxl__srm_callout_sendreply. It was in the switch
statement in the autogenerated _libxl_save_msgs_callout.c, and is now
in the handwritten libxl_dom.c.
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Shriram Rajagopalan <rshriram@cs.ubc.ca>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Clarify commit message.
Fix a misformatted CC in the commit message
Do not introduce a whitespace error in libxl_save_msgs_gen.pl
v2: Commit message mentions usleeps, not Remus, as motivation.
libxl_save_msgs_gen.pl has some trailing whitespace. This can make
git complain annoyingly. Remove the trailing whitespace.
(The file also contains tabs. Leave these as they are to avoid a big
and intrusive diff.)
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: New patch in v3 of the series.
Use fcntl F_SETLK, rather than writing our pid into a "lock" file.
That way if we crash we don't leave the lockfile lying about. Callers
now need to keep the fd for our lockfile. (We don't use flock because
we don't want anyone who inherits this fd across fork to end up with a
handle onto the lock.)
While we are here:
* Move the lockfile to /var/run/xen
* De-duplicate the calculation of the pathname
* Compute the buffer size for the pathname so that it will definitely
not overrun (and use the computed value everywhere)
* Fix various error handling bugs
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Shriram Rajagopalan <rshriram@cs.ubc.ca>
xc_suspend_evtchn_init expects to eat the first event on the xce. If
the xce is used for any other purpose then this can break. Document
this fact and rename the function to xc_suspend_evtchn_init_exclusive.
(I haven't checked the call sites for improper shared use of the xce.)
Provide a corresponding xc_suspend_evtchn_init_sane which doesn't try
to eat an event, and instead leaves the caller the ability to
demultiplex.
Also document that xc_await_suspend needs exclusive use of the xce.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
CC: Shriram Rajagopalan <rshriram@cs.ubc.ca>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Drop spurious addition of #include <assert.h>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v2: Fix commit message not to refer to libxl_ctx_alloc's gc, done earlier
Change type of port in evtchn_fd_callback to evtchn_port_or_error_t
Clarify comment about use of ctx->xce.
Fix typo in comment.
Special-case paths starting with '@' in libxl__xswait. Attempting to
read these from xenstore gives EINVAL. Callers waiting for (say)
@releaseDomain will be checking for some condition which can be
observed other than by looking at xenstore.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: New patch in this version of the series.
This is an ao utility for for conveniently doing a timed wait on
xenstore. It handles setting up and cancelling the timeout, and also
conveniently reads the key for you.
No callers yet in this patch.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v2: Fix doc comments to refer to correct rc values
Provide a more formal description of the semantics of
libxl__xs_transaction_start, _commit and _abort, in addition to the
usage pattern example.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
The comments for libxl__ev_time_isregistered and the corresponding
watch function even say that these should be const. Make it so.
Also fix libxl__ev_child_inuse and libxl__ev_spawn_inuse.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Change libxl__poller_init and libxl__poller__get to take a libxl__gc*
rather than a libxl_ctx*. The gc is not used for memory allocation
but simply to provide the standard local variable "gc" expected by the
convenience macros. Doing this makes the error logging more
convenient.
Hence, convert the logging calls to use the LOG* convenience macros.
And consequently, change the call sites, and the function bodies to
use CTX rather than ctx.
Also convert a call to malloc() (with error check) in
libxl__poller_get, to libxl__zalloc (no error check needed).
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Provide libxl__gc *gc for the second half of libxl_ctx_alloc.
(For the first half of the function, gc is in scope but set to NULL.)
This makes it possible to make gc-requiring calls. For example, it
makes error logging more convenient.
Make use of this by changing the logging calls to use the LOG*
convenience macros.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Update to versions from Debian Wheezy's autotools-dev package (20120608.1)
Adds ARM aarch64 support.
Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Jackson <ian.jackson@eu.citrix.com>
The previous patch (dda0b77d "tools/libxc: pass errno to callers of
xc_domain_save") did not jump to the proper label if the domU is HVM.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Reviewed-by: Olaf Hering <olaf@aepfle.de>
If libxl_pipe fails we leave the carefd locked, which translates to
the atfork lock remaining held. This would probably cause the process
to deadlock shortly afterwards.
Of course libxl_pipe is very unlikely to fail unless things are
already going very badly. This bug has not been observed anywhere as
far as we are aware.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
CC: George Dunlap <george.dunlap@eu.citrix.com>
This avoids the process being forked while a carefd is recorded in the
list but the actual fd has been closed. If that happened, a
subsequent libxl_postfork_child_noexec would attempt to close the fd
again. If we are lucky that results in a harmless warning; but if we
are unlucky the fd number has been reused and we close an unrelated
fd.
This race has not been observed anywhere as far as we are aware.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
CC: George Dunlap <george.dunlap@eu.citrix.com>
libxl_device_events_handler would fail to call AO_ABORT if it failed;
instead it would simply return rc. (This leaves the egc etc. from the
now-abolished stack frame potentially live, and leaves the ctx
locked.)
In xl, this is of no consequence, because xl will immediately exit in
this situation. This is very likely to be true in any other callers
(of which we don't know of any, anyway).
Coverity-ID: 1181840
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
CC: coverity@xenproject.org
This was part of
[PATCH v2 2/2] xl: Add "seize" option to PCI devices
but I accidentally applied v1.
Signed-off-by: George Dunlap <george.dunlap@eu.citrix.com>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
libxl_postfork_child_noexec would nestedly reaquire the non-recursive
"no_forking" mutex: atfork_lock uses it, as does sigchld_user_remove.
The result on Linux is that the process always deadlocks before
returning from this function.
This is used by xl's console child. So, the ultimate effect is that
xl with pygrub does not manage to connect to the pygrub console.
This behaviour was reported by Michael Young in Xen 4.4.0 RC5.
Also, the use of sigchld_user_remove in libxl_postfork_child_noexec is
not correct with SIGCHLD sharing. libxl_postfork_child_noexec is
documented to suffice if called only on one ctx. So deregistering the
ctx it's called on is not sufficient. Instead, we need a new approach
which discards the whole sigchld_user list and unconditionally removes
our SIGCHLD handler if we had one.
Prompted by this, clarify the semantics of
libxl_postfork_child_noexec. Specifically, expand on the meaning of
"quickly" by explaining what operations are not permitted; and
document the fact that the function doesn't reclaim the resources in
the ctxs.
And add a comment in libxl_postfork_child_noexec explaining the
internal concurrency situation.
This is an important bugfix. IMO the bug is a blocker for Xen 4.4.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Reported-by: M A Young <m.a.young@durham.ac.uk>
CC: Ian Campbell <Ian.Campbell@citrix.com>
CC: George Dunlap <george.dunlap@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Release-acked-by: George Dunlap <george.dunlap@eu.citrix.com>
(cherry picked from commit 5be1e95318)
QEMU_UPSTREAM_REVISION set back to master, to track the tip.
QEMU_TAG set to the specific changeset as is customary.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Having this here means we have to wait for a push gate pass, or fart
about which explicit pushes to master, to make an RC. The boot
messages for git builds already contain the git revision (as a
shorthash).
I will change the tarball creation checklist to seddery the -rc back
to -rcN, along with the other release-management-related changes (like
using an embedded copy of qemu).
If this patch meets with approval it should be thrown into the push
gate today, along with the patch for XSA-88, and then hopefully
nothing much else, so that we can get something suitable for making an
RC from by Friday.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Acked-by: Jan Beulich <jbeulich@suse.com>
Acked-by: George Dunlap <george.dunlap@eu.citrix.com>
Release-Acked-by: George Dunlap <george.dunlap@eu.citrix.com>
The test programs were getting the proper libxenlight.so on their link
line. Filter it out. Also change the soname of the test library to
match the real one, so that libxutil is satisfied with it.
Reported-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Olaf Hering <olaf@aepfle.de>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
We need to include the new TEST_PROG_OBJS and LIBXL_TEST_OBJS in the
appropriate dependencies. Otherwise we risk trying to build the test
program before gentypes is run.
Reported-by: Olaf Hering <olaf@aepfle.de>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Olaf Hering <olaf@aepfle.de>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
When a timeout has occurred, it is deregistered. However, we failed
to record this fact by updating etime->func. As a result,
libxl__ev_time_isregistered would say `true' for a timeout which has
already happened.
It is necessary to clear etime->func before the callback, because the
callback might want to reinstate the timeout, or might free the etime
(or its containing struct) entirely.
The results are that we might try to have the timeout occur again
(causing problems for the call site), and/or corrupt the timeout list.
This fixes the timedereg event system unit test.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Bring together the two places where etime->func() is called into a new
function time_occurs. For one call site this is pure code motion.
For the other the only semantic change is the introduction of a new
debugging message.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Test timeout deregistration idempotency. In the current tree this
test fails because ev->func is not cleared, meaning that a timeout
can be removed from the list more than once, corrupting the list.
It is necessary to use multiple timeouts to demonstrate this bug,
because removing the very same entry twice from a list in quick
succession, without modifying the list in other ways in between,
doesn't actually corrupt the list. (Since removing an entry from a
doubly-linked list just copies next and back from the disappearing
entry into its neighbours.)
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
We provide a new LIBXL_TESTS facility in the Makefile.
Also provide some helpful common routines for unit tests to use.
We don't want to put the weird test case entrypoints and the weird
test case code in the main libxl.so library. Symbol hiding prevents
us from simply directly linking the libxl_test_FOO.o in later. So
instead we provide a special library libxenlight_test.so which is used
only locally.
There are not yet any test cases defined; that will come in the next
patch.
Signed-off-by: Ian Jackson <ian.jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Use the new libxl__pipe_nonblock and _close functions, rather than
open coding the same logic. Now the pipe is nonblocking, which avoids
a race which could result in libxl deadlocking in a multithreaded
program.
Reported-by: Jim Fehlig <jfehlig@suse.com>
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Break out the pipe creation and destruction from the poller code
into two new functions libxl__pipe_nonblock and libxl__pipe_close.
Also change direct use of pipe() to libxl_pipe.
No overall functional difference other than minor differences in exact
log messages.
Also move libxl__self_pipe_wakeup and libxl__self_pipe_eatall into the
new pipe utilities section in libxl_event.c; this is pure code motion.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <ian.campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
--
v3: Mention that we switched pipe() -> libxl_pipe()
Previously, an application which had multiple libxl ctxs in multiple
threads, would have to itself plumb SIGCHLD through to each ctx.
Instead, permit multiple libxl ctxs to all share the SIGCHLD handler.
We keep a list of all the ctxs which are interested in SIGCHLD and
notify all of their self-pipes.
In more detail:
* sigchld_owner, the ctx* of the SIGCHLD owner, is replaced by
sigchld_users, a list of SIGCHLD users.
* Each ctx keeps track of whether it is on the users list, so that
libxl__sigchld_needed and libxl__sigchld_notneeded now instead of
idempotently installing and removing the handler, idempotently add
or remove the ctx from the list.
We ensure that we always have the SIGCHLD handler installed
iff the sigchld_users list is nonempty. To make this a bit
easier we make sigchld_installhandler_core and
sigchld_removehandler_core idempotent.
Specifically, the call sites for sigchld_installhandler_core and
sigchld_removehandler_core are updated to manipulate sigchld_users
and only call the install or remove functions as applicable.
* In the signal handler we walk the list of SIGCHLD users and write
to each of their self-pipes. That means that we need to arrange to
defer SIGCHLD when we are manipulating the list (to avoid the
signal handler interrupting our list manipulation); this is quite
tiresome to arrange.
The code as written will, on the first installation of the SIGCHLD
handler, firstly install the real handler, then immediately replace
it with the deferral handler. Doing it this way makes the code
clearer as it makes the SIGCHLD deferral machinery much more
self-contained (and hence easier to reason about).
* The first part of libxl__sigchld_notneeded is broken out into a new
function sigchld_user_remove (which is also needed during for
postfork). And of course that first part of the function is now
rather different, as explained above.
* sigchld_installhandler_core no longer takes the gc argument,
because it now deals with SIGCHLD for all ctxs.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
---
v3: Include bugfixes from "Fixup SIGCHLD sharing" patch:
* Use a mutex for defer_sigchld, to guard against concurrency
between the thread calling defer_sigchld and an instance of the
primary signal handler on another thread.
* libxl_sigchld_owner_libxl_always is incompatible with SIGCHLD
sharing. Document this correctly.
Fix "have have" error in comment.
Move removal of newly unused variables to previous patch.
v2.1: Provide feature test macro LIBXL_HAVE_SIGCHLD_SHARING
We are going to want introduce another call site in the final
substantive patch.
Pure code motion; no functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
---
v3: Remove now-unused variables from sigchld_installhandler_core
We are going to change these functions so that different libxl ctx's
can share a single SIGCHLD handler. Rename them now to a new name
which doesn't imply unconditional handler installation or removal.
Also note in the comments that they are idempotent.
No functional change.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
This is the feature test macro for libxl_childproc_sigchld_occurred
and libxl_sigchld_owner_libxl_always_selective_reap.
It is split out into this separate patch because: a single feature
test is sensible because we do not intend anyone to release or ship
libxl versions with one of these but not the other; but, the two
features are in separate patches for clarity; and, this just makes
reading the actual code easier.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Applications exist which want to use libxl in an event-driven mode but
which do not integrate child termination into their event system, but
instead reap all their own children synchronously.
In such an application libxl must own SIGCHLD but avoid reaping any
children that don't belong to libxl.
Provide libxl_sigchld_owner_libxl_always_selective_reap which has this
behaviour.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
---
v2: Document the new mode in the big "Subprocess handling" comment.
Applications exist which don't keep track of all their child processes
in a manner suitable for coherent dispatch of their termination. In
such a situation, nothing in the whole process may call wait, or
waitpid(-1,,). Doing so reaps processes belonging to other parts of
the application and there is then no way to deliver the exit status to
the right place.
To facilitate this, provide a facility for such an application to ask
libxl to call waitpid on each of its children individually.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
libxl_sigchld_owner_libxl ought to have been mentioned in the list of
options for chldowner. Since it's the default, move the description
of the its behaviour into the description of that option.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
Clarify that libxl_sigchld_owner_libxl causes libxl to reap all the
process's children, and clarify the wording of the description of
libxl_sigchld_owner_libxl_always.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
This is a simple error-handling wrapper for waitpid. We're going to
want to call waitpid somewhere else and this avoids some of the
duplication.
No functional change in this patch. (Technically, we used to check
chldmode_ours again in the EINTR case, and don't now, but that can't
have changed because we continuously hold the libxl ctx lock.)
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
There have been ABI/API changes in libxc. Bump its MAJOR (which
affets libxenguest et al too.)
There have been ABI changes in libxl. Bump its MAJOR.
(The API changes have been dealt with as we go along - there is
already a LIBXL_API_VERSION 0x040400.)
None of the other libraries have changed their interfaces. I have
verified this by building the tools and searching the dist/install
tree for files matching *.so.*. For each library that showed up, I
did this:
git-diff RELEASE-4.3.0..staging -- `find tools/FOO/ -name \*.h`
where FOO is the corresponding source directory.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
libxl_event.c:eventloop_iteration would pass the allocated pollfds
array size, rather than the used size, to poll (and to
afterpoll_internal).
The effect is that if the number of fds to poll on reduces, libxl will
poll on stale entries. Because of the way the return value from poll
is processed these stale entries are often harmless because any events
coming back from poll ignored by libxl. However, it could cause
malfunctions:
It could result in unwanted SIGTTIN/SIGTTOU/SIGPIPE, for example, if
the fd has been reused to refer to an object which can generate those
signals. Alternatively, it could result in libxl spinning if the
stale entry refers to an fd which happens now to be ready for the
previously-requested operation.
I have tested this with a localhost migration and inspected the strace
output.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Cc: Jim Fehlig <jfehlig@suse.com>
Cc: Ian Campbell <Ian.Campbell@citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
As Ian Campbell writes in http://bugs.xenproject.org/xen/bug/30:
There are two mechanisms by which a suspend can be aborted and the
original domain resumed.
The older method is that the toolstack resets a bunch of state (see
tools/python/xen/xend/XendDomainInfo.py resumeDomain) and then
restarts the domain. The domain will see HYPERVISOR_suspend return 0
and will continue without any realisation that it is actually
running in the original domain and not in a new one. This method is
supposed to be implemented by libxl_domain_resume(suspend_cancel=0)
but it is not.
The other method is newer and in this case the toolstack arranges
that HYPERVISOR_suspend returns SUSPEND_CANCEL and restarts it. The
domain will observe this and realise that it has been restarted in
the same domain and will behave accordingly. This method is
implemented, correctly AFAIK, by
libxl_domain_resume(suspend_cancel=1).
Attempting to use the old method without doing all of the work simply
causes the guest to crash. Implementing the work required for old
method, or for checking that domains actually support the new method,
is not feasible at this stage of the 4.4 release.
So, always use the new method, without regard to the declarations of
support by the guest. This is a strict improvement: guests which do
in fact support the new method will work, whereas ones which don't are
no worse off.
There are two call sites of libxl_domain_resume that need fixing, both
in the migration error path.
With this change I observe a correct and successful resumption of a
Debian wheezy guest with a Linux 3.4.70 kernel after a migration
attempt which I arranged to fail by nobbling the block hotplug script.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <Ian.Campbell@citrix.com>
CC: konrad.wilk@oracle.com
CC: David Vrabel <david.vrabel@citrix.com>
CC: Boris Ostrovsky <boris.ostrovsky@oracle.com>
Provide semantic documentation for how the libxc calls relate to the
hypervisor interface, and how they are to be used.
Also document the bug (present at least in Linux 3.12) that setting
the evtchn fd to nonblocking doesn't in fact make xc_evtchn_pending
nonblocking, and describe the appropriate workaround.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
CC: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
CC: Jan Beulich <JBeulich@suse.com>
CC: Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>
Document the event channel protocol in xenstore-paths.markdown, in the
section for ~/device/suspend/event-channel.
Protocol reverse-engineered from commentary and commit messages of
4539594d46 Add facility to get notification of domain suspend ...
17636f47a4 Teach xc_save to use event-channel-based ...
and implementations in
xc_save (current version)
libxl (current version)
linux-2.6.18-xen (mercurial 1241:2993033a77ca)
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
CC: Keir Fraser <keir@xen.org>
CC: Shriram Rajagopalan <rshriram@cs.ubc.ca>
Previously, this function would leak the temporary return from xs_read for
handle and mac address. Fix both of these and the rest of the error handling.
This requires changing its return type and fixing the callers.
Introduce here a READ_BACKEND macro to make the code less repetitive.
Coverity ID: 1055886
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
[ ijc -- spell out what the leaks were in the commit message ]
Update the comment to clarify that libxl_sigchld_owner_libxl_always
implies having only one libxl_ctx, and that
libxl_sigchld_owner_mainloop requires one call to
libxl_childproc_exited per ctx.
Signed-off-by: Ian Jackson <Ian.Jackson@eu.citrix.com>
Acked-by: Ian Campbell <ian.campbell@citrix.com>
Ian Jackson