xen.git
3 years agoCommit patch queue (exported by git-debrebase)
Hans van Kranenburg [Fri, 24 Feb 2023 17:10:30 +0000 (18:10 +0100)]
Commit patch queue (exported by git-debrebase)

[git-debrebase make-patches: export and commit patches]

3 years agoxen/arch/x86: make objdump output user locale agnostic
Maximilian Engelhardt [Thu, 9 Dec 2021 23:23:30 +0000 (00:23 +0100)]
xen/arch/x86: make objdump output user locale agnostic

The objdump output is fed to grep, so make sure it doesn't change with
different user locales and break the grep parsing.
This problem was identified while updating xen in Debian and the fix is
needed for generating reproducible builds in varying environments.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agogive meaningful error message if qemu device model is unavailable
Michael Tokarev [Sun, 24 Apr 2022 09:26:38 +0000 (12:26 +0300)]
give meaningful error message if qemu device model is unavailable

There's no sense to switch to qemu-xen-traditional device model
if that one is not enabled in the first place. This way we'll
have a chance later to print a message suggesting to install the
missing qemu package if we *actually* need qemu for the device model.

3 years agodocs: set date to SOURCE_DATE_EPOCH if available
Maximilian Engelhardt [Fri, 18 Dec 2020 20:42:35 +0000 (21:42 +0100)]
docs: set date to SOURCE_DATE_EPOCH if available

Use the solution described in [1] to replace the call to the 'date'
command with a version that uses SOURCE_DATE_EPOCH if available. This
is needed for reproducible builds.

[1] https://reproducible-builds.org/docs/source-date-epoch/

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
[Hans van Kranenburg]
Note: this patch is submitted upstream but not committed yet. We
expect that it gets in. Otherwise, we don't wait and already have it
here because I want to have the reproducible build work completed.

3 years agotools: don't build/ship xenmon
Hans van Kranenburg [Sat, 5 Sep 2020 20:43:19 +0000 (22:43 +0200)]
tools: don't build/ship xenmon

This is something that hasn't been touched (except for making it Python
3 compatible, which failed) since 2007. Don't build or ship it.

    -# xenmon
      File "/usr/sbin/xenmon", line 680
stop_cmd = "/usr/bin/pkill -INT -z global xenbaked"
    TabError: inconsistent use of tabs and spaces in indentation

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agotools/xl/bash-completion: also complete 'xen'
Hans van Kranenburg [Sun, 10 Feb 2019 17:26:45 +0000 (18:26 +0100)]
tools/xl/bash-completion: also complete 'xen'

We have the `xen` alias for xl in Debian, since in the past it was a
command that could execute either xl or xm.

Now, it always does xl, so, complete the same stuff for it as we have
for xl.

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
[git-debrebase split: mixed commit: upstream part]

3 years agopygrub: Specify -rpath LIBEXEC_LIB when building fsimage.so
Ian Jackson [Fri, 22 Feb 2019 12:24:35 +0000 (12:24 +0000)]
pygrub: Specify -rpath LIBEXEC_LIB when building fsimage.so

If LIBEXEC_LIB is not on the default linker search path, the python
fsimage.so module fails to find libfsimage.so.

Add the relevant directory to the rpath explicitly.

(This situation occurs in the Debian package, where
--with-libexec-libdir is used to put each Xen version's libraries and
utilities in their own directory, to allow them to be coinstalled.)

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agopygrub: Set sys.path
Bastian Blank [Sat, 5 Jul 2014 09:47:01 +0000 (11:47 +0200)]
pygrub: Set sys.path

We install libfsimage in a non-standard path for Reasons.
(See debian/rules.)

This patch was originally part of `tools-pygrub-prefix.diff'
(eg commit 51657319be54) and included changes to the Makefile to
change the installation arrangements (we do that part in the rules now
since that is a lot less prone to conflicts when we update) and to
shared library rpath (which is now done in a separate patch).

(Commit message rewritten by Ian Jackson.)

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
squash! pygrub: Set sys.path and rpath

3 years agohotplug-common: Do not adjust LD_LIBRARY_PATH
Ian Jackson [Thu, 21 Feb 2019 16:05:40 +0000 (16:05 +0000)]
hotplug-common: Do not adjust LD_LIBRARY_PATH

This is in the upstream script because on non-Debian systems, the
default install locations in /usr/local/lib might not be on the linker
path, and as a result the hotplug scripts would break.

A reason we might need it in Debian is our multiple version
coinstallation scheme.  However, the hotplug scripts all call the
utilities via the wrappers, and the binaries are configured to load
from the right place anyway.

This setting is an annoyance because it requires libdir, which is an
arch-specific path but comes from a file we want to put in
xen-utils-common, an arch:all package.

So drop this setting.

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agosysconfig.xencommons.in: Strip and debianize
Hans van Kranenburg [Sat, 9 Feb 2019 16:27:26 +0000 (17:27 +0100)]
sysconfig.xencommons.in: Strip and debianize

Strip all options that are for stuff we don't ship, which is 1)
xenstored as stubdom and 2) the new options for oom score and open file
descriptor limit, which would not have any effect, because we're
shipping different init scripts... :|

It seems useful to give the user the option to revert to xenstored
instead of the default oxenstored if they really want.

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
Acked-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
3 years agot/h/L/vif-common.sh: disable handle_iptable
Hans van Kranenburg [Thu, 3 Jan 2019 23:35:45 +0000 (00:35 +0100)]
t/h/L/vif-common.sh: disable handle_iptable

Also see Debian bug #894013. The current attempt at providing
anti-spoofing rules results in a situation that does not have any
effect. Also note that forwarding bridged traffic to iptables is not
enabled by default, and that for openvswitch users it does not make any
sense.

So, stop cluttering the live iptables ruleset.

This functionality seems to be introduced before 2004 and since then it
has never got some additional love.

It would be nice to have a proper discussion upstream about how Xen
could provide some anti mac/ip spoofing in the dom0. It does not seem to
be a trivial thing to do, since it requires having quite some knowledge
about what the domU is allowed to do or not (e.g. a domU can be a
router...).

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agodocs/man/xen-vbd-interface.7: Provide properly-formatted NAME section
Ian Jackson [Fri, 12 Oct 2018 16:56:56 +0000 (17:56 +0100)]
docs/man/xen-vbd-interface.7: Provide properly-formatted NAME section

This manpage was omitted from
   docs/man: Provide properly-formatted NAME sections
because I was previously building with markdown not installed.

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agoshim: Provide separate install-shim target
Ian Jackson [Fri, 12 Oct 2018 17:17:10 +0000 (17:17 +0000)]
shim: Provide separate install-shim target

When building on a 32-bit userland, the user wants to build 32-bit
tools and a 64-bit hypervisor.  This involves setting XEN_TARGET_ARCH
to different values for the tools build and the hypervisor build.

So the user must invoke the tools build and the hypervisor build
separately.

However, although the shim is done by the tools/firmware Makefile, its
bitness needs to be the same as the hypervisor, not the same as the
tools.  When run with XEN_TARGET_ARCH=x86_32, it it skipped, which is
wrong.

So the user must invoke the shim build separately.  This can be done
with
   make -C tools/firmware/xen-dir XEN_TARGET_ARCH=x86_64

However, tools/firmware/xen-dir has no `install' target.  The
installation of all `firmware' is done in tools/firmware/Makefile.  It
might be possible to fix this, but it is not trivial.  For example,
the definitions of INST_DIR and DEBG_DIR would need to be copied, as
would an appropriate $(INSTALL_DIR) call.

For now, provide an `install-shim' target in tools/firmware/Makefile.

This has to be called from `install' of course.  We can't make it
a dependency of `install' because it might be run before `all' has
completed.  We could make it depend on a `shim' target but such
a target is nearly impossible to write because everything is done by
the inflexible subdir-$@ machinery.

The overally result of this patch is that existing make invocations
work as before.  But additionally, the user can say
  make -C tools/firmware install-shim XEN_TARGET_ARCH=x86_64
to install the shim.  The user must have built it already.
Unlike the build rune, this install-rune is properly conditional
so it is OK to call on ARM.

What a mess.

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
3 years agoconfig/Tools.mk.in: Respect caller's CONFIG_PV_SHIM
Ian Jackson [Fri, 12 Oct 2018 16:00:16 +0000 (16:00 +0000)]
config/Tools.mk.in: Respect caller's CONFIG_PV_SHIM

This makes it easier to disable the shim build.  (In Debian we need to
build the shim separately because it needs different compiler flags).

Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
[ Hans: adjust from tools/firmware/Makefile to config/Tools.mk.in to
follow changes that happened in 8845155c83 ("pvshim: make PV shim build
selectable from configure") ]
Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years ago.gitignore: Add configure output which we always delete and regenerate
Ian Jackson [Fri, 5 Oct 2018 17:05:48 +0000 (18:05 +0100)]
.gitignore: Add configure output which we always delete and regenerate

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agoautoconf: Provide libexec_libdir_suffix
Ian Jackson [Wed, 3 Oct 2018 15:25:58 +0000 (16:25 +0100)]
autoconf: Provide libexec_libdir_suffix

This is going to be used to put libfsimage.so into a path containing
the multiarch triplet.

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agotools-libfsimage-prefix.diff
Hans van Kranenburg [Mon, 25 May 2020 15:08:18 +0000 (17:08 +0200)]
tools-libfsimage-prefix.diff

\o/

3 years agoDo not build the instruction emulator
Ian Jackson [Thu, 20 Sep 2018 17:10:14 +0000 (18:10 +0100)]
Do not build the instruction emulator

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agoRemove static solaris support from pygrub
Bastian Blank [Sat, 5 Jul 2014 09:47:29 +0000 (11:47 +0200)]
Remove static solaris support from pygrub

Patch-Name: tools-pygrub-remove-static-solaris-support

Gbp-Pq: Topic misc
Gbp-Pq: Name tools-pygrub-remove-static-solaris-support

3 years agoDo not ship COPYING into /usr/include
Bastian Blank [Sat, 5 Jul 2014 09:47:30 +0000 (11:47 +0200)]
Do not ship COPYING into /usr/include

This is not wanted in Debian.  COPYING ends up in
/usr/share/doc/xen-*copyright.

Patch-Name: tools-include-no-COPYING.diff

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agoconfig-prefix.diff
Bastian Blank [Sat, 5 Jul 2014 09:46:45 +0000 (11:46 +0200)]
config-prefix.diff

Patch-Name: config-prefix.diff

Gbp-Pq: Topic prefix-abiname
Gbp-Pq: Name config-prefix.diff

3 years agoDisplay Debian package version in hypervisor log
Bastian Blank [Sat, 5 Jul 2014 09:46:43 +0000 (11:46 +0200)]
Display Debian package version in hypervisor log

During hypervisor boot, disable the banner and nicely display the xen
version as well as the Maintainer address from debian/control.

For this to work the DEB_VERSION and DEB_MAINTAINER variables needs to
be set by debian/rules.

Original patch by Bastian Blank <waldi@debian.org>
Modified by
Hans van Kranenburg <hans@knorrie.org>
Maximilian Engelhardt <maxi@daemonizer.de>

3 years agoDelete configure output
Ian Jackson [Wed, 19 Sep 2018 15:53:22 +0000 (16:53 +0100)]
Delete configure output

These autogenerated files are not useful in Debian; dh_autoreconf will
regenerate them.

If this patch does not apply when rebasing, you can simply delete the
files again.

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agoDelete config.sub and config.guess
Ian Jackson [Wed, 19 Sep 2018 15:45:49 +0000 (16:45 +0100)]
Delete config.sub and config.guess

dh_autoreconf will provide these back.

If this patch does not apply when rebasing, you can simply delete the
files again.

Signed-off-by: Ian Jackson <ian.jackson@citrix.com>
3 years agodebian/changelog: finish 4.17.0+46-gaaf74a532c-1
Hans van Kranenburg [Fri, 24 Feb 2023 17:08:07 +0000 (18:08 +0100)]
debian/changelog: finish 4.17.0+46-gaaf74a532c-1

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agodebian/changelog: Remove duplicate 'Note that'
Hans van Kranenburg [Fri, 10 Feb 2023 12:59:21 +0000 (13:59 +0100)]
debian/changelog: Remove duplicate 'Note that'

This was already included in the changelog for 4.17.0-1 :(

Signed-off-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com>
3 years agodebian/changelog: Fix bug number typo.
Hans van Kranenburg [Fri, 10 Feb 2023 12:57:58 +0000 (13:57 +0100)]
debian/changelog: Fix bug number typo.

00:30 < Maxi[m]> Knorrie: I just noticed, the "(Closes: #102983)" from
   our changelog is missing a 0 at the end.
00:30 -zwiebelbot:#debian-xen- Debian#102983:
   quantlib_0.1.9-1(unstable): please add build-depends -
   https://bugs.debian.org/102983
00:31 < Maxi[m]> The correct bug number is #1029830

Oops. We will have to set it do done manually.

Signed-off-by: Hans van Kranenburg <hans.van.kranenburg@mendix.com>
3 years agodebian/shuffle-boot-files: fix typo
Hans van Kranenburg [Sat, 4 Feb 2023 16:57:23 +0000 (17:57 +0100)]
debian/shuffle-boot-files: fix typo

The tree picture changed, but I didn't correct the names in the text.
:-)

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agoUpdate changelog for new upstream 4.17.0+46-gaaf74a532c
Hans van Kranenburg [Fri, 24 Feb 2023 17:06:42 +0000 (18:06 +0100)]
Update changelog for new upstream 4.17.0+46-gaaf74a532c

[git-debrebase changelog: new upstream 4.17.0+46-gaaf74a532c]

3 years agoUpdate to upstream 4.17.0+46-gaaf74a532c
Hans van Kranenburg [Fri, 24 Feb 2023 17:06:42 +0000 (18:06 +0100)]
Update to upstream 4.17.0+46-gaaf74a532c

[git-debrebase anchor: new upstream 4.17.0+46-gaaf74a532c, merge]

3 years agod/changelog: finish 4.17.0+24-g2f8851c37f-2
Hans van Kranenburg [Mon, 6 Feb 2023 13:41:15 +0000 (14:41 +0100)]
d/changelog: finish 4.17.0+24-g2f8851c37f-2

3 years agochangelog: Prepare for upload to experimental
Ian Jackson [Sun, 5 Feb 2023 13:08:06 +0000 (13:08 +0000)]
changelog: Prepare for upload to experimental

3 years agoautomation: Remove clang-8 from Debian unstable container
Anthony PERARD [Tue, 21 Feb 2023 16:55:38 +0000 (16:55 +0000)]
automation: Remove clang-8 from Debian unstable container

First, apt complain that it isn't the right way to add keys anymore,
but hopefully that's just a warning.

Second, we can't install clang-8:
The following packages have unmet dependencies:
 clang-8 : Depends: libstdc++-8-dev but it is not installable
           Depends: libgcc-8-dev but it is not installable
           Depends: libobjc-8-dev but it is not installable
           Recommends: llvm-8-dev but it is not going to be installed
           Recommends: libomp-8-dev but it is not going to be installed
 libllvm8 : Depends: libffi7 (>= 3.3~20180313) but it is not installable
E: Unable to correct problems, you have held broken packages.

clang on Debian unstable is now version 14.0.6.

Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
(cherry picked from commit a6b1e2b80fe2053b1c9c9843fb086a668513ea36)

3 years agox86/spec-ctrl: Mitigate Cross-Thread Return Address Predictions
Andrew Cooper [Thu, 8 Sep 2022 20:27:58 +0000 (21:27 +0100)]
x86/spec-ctrl: Mitigate Cross-Thread Return Address Predictions

This is XSA-426 / CVE-2022-27672

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
(cherry picked from commit 63305e5392ec2d17b85e7996a97462744425db80)

3 years agotools/ocaml/libs: Fix memory/resource leaks with caml_alloc_custom()
Andrew Cooper [Wed, 1 Feb 2023 11:27:42 +0000 (11:27 +0000)]
tools/ocaml/libs: Fix memory/resource leaks with caml_alloc_custom()

All caml_alloc_*() functions can throw exceptions, and longjump out of
context.  If this happens, we leak the xch/xce handle.

Reorder the logic to allocate the the Ocaml object first.

Fixes: 8b3c06a3e545 ("tools/ocaml/xenctrl: OCaml 5 support, fix use-after-free")
Fixes: 22d5affdf0ce ("tools/ocaml/evtchn: OCaml 5 support, fix potential resource leak")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit d69ccf52ad467ccc22029172a8e61dc621187889)

3 years agotools/ocaml/xc: Don't reference Abstract_Tag objects with the GC lock released
Andrew Cooper [Tue, 31 Jan 2023 17:19:30 +0000 (17:19 +0000)]
tools/ocaml/xc: Don't reference Abstract_Tag objects with the GC lock released

The intf->{addr,len} references in the xc_map_foreign_range() call are unsafe.
From the manual:

  https://ocaml.org/manual/intfc.html#ss:parallel-execution-long-running-c-code

"After caml_release_runtime_system() was called and until
caml_acquire_runtime_system() is called, the C code must not access any OCaml
data, nor call any function of the run-time system, nor call back into OCaml
code."

More than what the manual says, the intf pointer is (potentially) invalidated
by caml_enter_blocking_section() if another thread happens to perform garbage
collection at just the right (wrong) moment.

Rewrite the logic.  There's no need to stash data in the Ocaml object until
the success path at the very end.

Fixes: 8b7ce06a2d34 ("ocaml: Add XC bindings.")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 9e7c74e6f9fd2e44df1212643b80af9032b45b07)

3 years agotools/ocaml/xc: Fix binding for xc_domain_assign_device()
Edwin Török [Thu, 12 Jan 2023 11:38:38 +0000 (11:38 +0000)]
tools/ocaml/xc: Fix binding for xc_domain_assign_device()

The patch adding this binding was plain broken, and unreviewed.  It modified
the C stub to add a 4th parameter without an equivalent adjustment in the
Ocaml side of the bindings.

In 64bit builds, this causes us to dereference whatever dead value is in %rcx
when trying to interpret the rflags parameter.

This has gone unnoticed because Xapi doesn't use this binding (it has its
own), but unbreak the binding by passing RDM_RELAXED unconditionally for
now (matching the libxl default behaviour).

Fixes: 9b34056cb4 ("tools: extend xc_assign_device() to support rdm reservation policy")
Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 4250683842104f02996428f93927a035c8e19266)

3 years agotools/ocaml/evtchn: Don't reference Custom objects with the GC lock released
Edwin Török [Thu, 12 Jan 2023 17:48:29 +0000 (17:48 +0000)]
tools/ocaml/evtchn: Don't reference Custom objects with the GC lock released

The modification to the _H() macro for Ocaml 5 support introduced a subtle
bug.  From the manual:

  https://ocaml.org/manual/intfc.html#ss:parallel-execution-long-running-c-code

"After caml_release_runtime_system() was called and until
caml_acquire_runtime_system() is called, the C code must not access any OCaml
data, nor call any function of the run-time system, nor call back into OCaml
code."

Previously, the value was a naked C pointer, so dereferencing it wasn't
"accessing any Ocaml data", but the fix to avoid naked C pointers added a
layer of indirection through an Ocaml Custom object, meaning that the common
pattern of using _H() in a blocking section is unsafe.

In order to fix:

 * Drop the _H() macro and replace it with a static inline xce_of_val().
 * Opencode the assignment into Data_custom_val() in the two constructors.
 * Rename "value xce" parameters to "value xce_val" so we can consistently
   have "xenevtchn_handle *xce" on the stack, and obtain the pointer with the
   GC lock still held.

Fixes: 22d5affdf0ce ("tools/ocaml/evtchn: OCaml 5 support, fix potential resource leak")
Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 2636d8ff7a670c4d2485757dbe966e36c259a960)

3 years agotools/ocaml/libs: Allocate the correct amount of memory for Abstract_tag
Andrew Cooper [Tue, 31 Jan 2023 10:59:42 +0000 (10:59 +0000)]
tools/ocaml/libs: Allocate the correct amount of memory for Abstract_tag

caml_alloc() takes units of Wsize (word size), not bytes.  As a consequence,
we're allocating 4 or 8 times too much memory.

Ocaml has a helper, Wsize_bsize(), but it truncates cases which aren't an
exact multiple.  Use a BUILD_BUG_ON() to cover the potential for truncation,
as there's no rounding-up form of the helper.

Fixes: 8b7ce06a2d34 ("ocaml: Add XC bindings.")
Fixes: d3e649277a13 ("ocaml: add mmap bindings implementation.")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 36eb2de31b6ecb8787698fb1a701bd708c8971b2)

3 years agotools/ocaml/libs: Don't declare stubs as taking void
Edwin Török [Thu, 12 Jan 2023 11:28:29 +0000 (11:28 +0000)]
tools/ocaml/libs: Don't declare stubs as taking void

There is no such thing as an Ocaml function (C stub or otherwise) taking no
parameters.  In the absence of any other parameters, unit is still passed.

This doesn't explode with any ABI we care about, but would malfunction for an
ABI environment such as stdcall.

Fixes: c3afd398ba7f ("ocaml: Add XS bindings.")
Fixes: 8b7ce06a2d34 ("ocaml: Add XC bindings.")
Signed-off-by: Edwin Török <edwin.torok@cloud.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit ff8b560be80b9211c303d74df7e4b3921d2bb8ca)

3 years agotools/oxenstored: validate config file before live update
Edwin Török [Tue, 11 May 2021 15:56:50 +0000 (15:56 +0000)]
tools/oxenstored: validate config file before live update

The configuration file can contain typos or various errors that could prevent
live update from succeeding (e.g. a flag only valid on a different version).
Unknown entries in the config file would be ignored on startup normally,
add a strict --config-test that live-update can use to check that the config file
is valid *for the new binary*.

For compatibility with running old code during live update recognize
--live --help as an equivalent to --config-test.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit e6f07052ce4a0f0b7d4dc522d87465efb2d9ee86)

3 years agotools/ocaml/xb: Drop Xs_ring.write
Edwin Török [Fri, 16 Dec 2022 18:25:20 +0000 (18:25 +0000)]
tools/ocaml/xb: Drop Xs_ring.write

This function is unusued (only Xs_ring.write_substring is used), and the
bytes/string conversion here is backwards: the C stub implements the bytes
version and then we use a Bytes.unsafe_of_string to convert a string into
bytes.

However the operation here really is read-only: we read from the string and
write it to the ring, so the C stub should implement the read-only string
version, and if needed we could use Bytes.unsafe_to_string to be able to send
'bytes'. However that is not necessary as the 'bytes' version is dropped above.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 01f139215e678c2dc7d4bb3f9f2777069bb1b091)

3 years agotools/ocaml/xb,mmap: Use Data_abstract_val wrapper
Edwin Török [Fri, 16 Dec 2022 18:25:10 +0000 (18:25 +0000)]
tools/ocaml/xb,mmap: Use Data_abstract_val wrapper

This is not strictly necessary since it is essentially a no-op currently: a
cast to void * and value *, even in OCaml 5.0.

However it does make it clearer that what we have here is not a regular OCaml
value, but one allocated with Abstract_tag or Custom_tag, and follows the
example from the manual more closely:
https://v2.ocaml.org/manual/intfc.html#ss:c-outside-head

It also makes it clearer that these modules have been reviewed for
compat with OCaml 5.0.

We cannot use OCaml finalizers here, because we want exact control over when
to unmap these pages from remote domains.

No functional change.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit d2ccc637111d6dbcf808aaffeec7a46f0b1e1c81)

3 years agotools/ocaml/xenctrl: Use larger chunksize in domain_getinfolist
Edwin Török [Tue, 1 Nov 2022 17:59:17 +0000 (17:59 +0000)]
tools/ocaml/xenctrl: Use larger chunksize in domain_getinfolist

domain_getinfolist() is quadratic with the number of domains, because of the
behaviour of the underlying hypercall.  Nevertheless, getting domain info in
blocks of 1024 is far more efficient than blocks of 2.

In a scalability testing scenario with ~1000 VMs, a combination of this and
the previous change takes xenopsd's wallclock time in domain_getinfolist()
down from 88% to 0.02%

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Tested-by: Pau Ruiz Safont <pau.safont@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 95db09b1b154fb72fad861815ceae1f3fa49fc4e)

3 years agotools/ocaml/xenctrl: Make domain_getinfolist tail recursive
Edwin Török [Tue, 1 Nov 2022 17:59:16 +0000 (17:59 +0000)]
tools/ocaml/xenctrl: Make domain_getinfolist tail recursive

domain_getinfolist() is quadratic with the number of domains, because of the
behaviour of the underlying hypercall.  xenopsd was further observed to be
wasting excessive quantites of time manipulating the list of already-obtained
domains.

Implement a tail recursive `rev_concat` equivalent to `concat |> rev`, and use
it instead of calling `@` multiple times.

An incidental benefit is that the list of domains will now be in domid order,
instead of having pairs of 2 domains changing direction every time.

In a scalability testing scenario with ~1000 VMs, a combination of this and
the subsequent change takes xenopsd's wallclock time in domain_getinfolist()
down from 88% to 0.02%

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Tested-by: Pau Ruiz Safont <pau.safont@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit c3b6be714c64aa62b56d0bce96f4b6a10b5c2078)

3 years agolibxl: fix guest kexec - skip cpuid policy
Jason Andryuk [Tue, 7 Feb 2023 16:01:49 +0000 (17:01 +0100)]
libxl: fix guest kexec - skip cpuid policy

When a domain performs a kexec (soft reset), libxl__build_pre() is
called with the existing domid.  Calling libxl__cpuid_legacy() on the
existing domain fails since the cpuid policy has already been set, and
the guest isn't rebuilt and doesn't kexec.

xc: error: Failed to set d1's policy (err leaf 0xffffffff, subleaf 0xffffffff, msr 0xffffffff) (17 = File exists): Internal error
libxl: error: libxl_cpuid.c:494:libxl__cpuid_legacy: Domain 1:Failed to apply CPUID policy: File exists
libxl: error: libxl_create.c:1641:domcreate_rebuild_done: Domain 1:cannot (re-)build domain: -3
libxl: error: libxl_xshelp.c:201:libxl__xs_read_mandatory: xenstore read failed: `/libxl/1/type': No such file or directory
libxl: warning: libxl_dom.c:49:libxl__domain_type: unable to get domain type for domid=1, assuming HVM

During a soft_reset, skip calling libxl__cpuid_legacy() to avoid the
issue.  Before commit 34990446ca91, the libxl__cpuid_legacy() failure
would have been ignored, so kexec would continue.

Fixes: 34990446ca91 ("libxl: don't ignore the return value from xc_cpuid_apply_policy")
Signed-off-by: Jason Andryuk <jandryuk@gmail.com>
Reviewed-by: Anthony PERARD <anthony.perard@citrix.com>
master commit: 1e454c2b5b1172e0fc7457e411ebaba61db8fc87
master date: 2023-01-26 10:58:23 +0100

3 years agons16550: fix an incorrect assignment to uart->io_size
Ayan Kumar Halder [Tue, 7 Feb 2023 16:00:47 +0000 (17:00 +0100)]
ns16550: fix an incorrect assignment to uart->io_size

uart->io_size represents the size in bytes. Thus, when serial_port.bit_width
is assigned to it, it should be converted to size in bytes.

Fixes: 17b516196c ("ns16550: add ACPI support for ARM only")
Reported-by: Jan Beulich <jbeulich@suse.com>
Signed-off-by: Ayan Kumar Halder <ayan.kumar.halder@amd.com>
Reviewed-by: Stefano Stabellini <sstabellini@kernel.org>
master commit: 352c89f72ddb67b8d9d4e492203f8c77f85c8df1
master date: 2023-01-24 16:54:38 +0100

3 years agobuild: fix building flask headers before descending in flask/ss/
Anthony PERARD [Tue, 7 Feb 2023 16:00:29 +0000 (17:00 +0100)]
build: fix building flask headers before descending in flask/ss/

Unfortunatly, adding prerequisite to "$(obj)/ss/built_in.o" doesn't
work because we have "$(obj)/%/built_in.o: $(obj)/% ;" in Rules.mk.
So, make is allow to try to build objects in "xsm/flask/ss/" before
generating the headers.

Adding a prerequisite on "$(obj)/ss" instead will fix the issue as
that's the target used to run make in this subdirectory.

Unfortunatly, that target is also used when running `make clean`, so
we want to ignore it in this case. $(MAKECMDGOALS) can't be used in
this case as it is empty, but we can guess which operation is done by
looking at the list of loaded makefiles.

Fixes: 7a3bcd2babcc ("build: build everything from the root dir, use obj=$subdir")
Reported-by: "Daniel P. Smith" <dpsmith@apertussolutions.com>
Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Acked-by: Daniel P. Smith <dpsmith@apertussolutions.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
master commit: d60324d8af9404014cfcc37bba09e9facfd02fcf
master date: 2023-01-23 15:03:58 +0100

3 years agox86/shadow: fix PAE check for top-level table unshadowing
Jan Beulich [Tue, 7 Feb 2023 15:59:54 +0000 (16:59 +0100)]
x86/shadow: fix PAE check for top-level table unshadowing

Clearly within the for_each_vcpu() the vCPU of this loop is meant, not
the (loop invariant) one the fault occurred on.

Fixes: 3d5e6a3ff383 ("x86 hvm: implement HVMOP_pagetable_dying")
Fixes: ef3b0d8d2c39 ("x86/shadow: shadow_table[] needs only one entry for PV-only configs")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Andrew Cooper <andrew.cooper3@citrix.com>
master commit: f8fdceefbb1193ec81667eb40b83bc525cb71204
master date: 2023-01-20 09:23:42 +0100

3 years agox86/vmx: Support for CPUs without model-specific LBR
Andrew Cooper [Tue, 7 Feb 2023 15:59:14 +0000 (16:59 +0100)]
x86/vmx: Support for CPUs without model-specific LBR

Ice Lake (server at least) has both architectural LBR and model-specific LBR.
Sapphire Rapids does not have model-specific LBR at all.  I.e. On SPR and
later, model_specific_lbr will always be NULL, so we must make changes to
avoid reliably hitting the domain_crash().

The Arch LBR spec states that CPUs without model-specific LBR implement
MSR_DBG_CTL.LBR by discarding writes and always returning 0.

Do this for any CPU for which we lack model-specific LBR information.

Adjust the now-stale comment, now that the Arch LBR spec has created a way to
signal "no model specific LBR" to guests.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
master commit: 3edca52ce736297d7fcf293860cd94ef62638052
master date: 2023-01-12 18:42:00 +0000

3 years agox86/vmx: Calculate model-specific LBRs once at start of day
Andrew Cooper [Tue, 7 Feb 2023 15:58:25 +0000 (16:58 +0100)]
x86/vmx: Calculate model-specific LBRs once at start of day

There is no point repeating this calculation at runtime, especially as it is
in the fallback path of the WRSMR/RDMSR handlers.

Move the infrastructure higher in vmx.c to avoid forward declarations,
renaming last_branch_msr_get() to get_model_specific_lbr() to highlight that
these are model-specific only.

No practical change.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Kevin Tian <kevin.tian@intel.com>
master commit: e94af0d58f86c3a914b9cbbf4d9ed3d43b974771
master date: 2023-01-12 18:42:00 +0000

3 years agoinclude/compat: produce stubs for headers not otherwise generated
Jan Beulich [Tue, 7 Feb 2023 15:57:52 +0000 (16:57 +0100)]
include/compat: produce stubs for headers not otherwise generated

Public headers can include other public headers. Such interdependencies
are retained in their compat counterparts. Since some compat headers are
generated only in certain configurations, the referenced headers still
need to exist. The lack thereof was observed with hvm/hvm_op.h needing
trace.h, where generation of the latter depends on TRACEBUFFER=y. Make
empty stubs in such cases (as generating the extra headers is relatively
slow and hence better to avoid). Changes to .config and incrementally
(re-)building is covered by the respective .*.cmd then no longer
matching the command to be used, resulting in the necessary re-creation
of the (possibly stub) header.

Reported-by: Andrew Cooper <andrew.cooper3@citrix.com>
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Anthony PERARD <anthony.perard@citrix.com>
master commit: 6bec713f871f21c6254a5783c1e39867ea828256
master date: 2023-01-12 16:17:54 +0100

3 years agotools: Fix build with recent QEMU, use "--enable-trace-backends"
Anthony PERARD [Tue, 7 Feb 2023 15:57:22 +0000 (16:57 +0100)]
tools: Fix build with recent QEMU, use "--enable-trace-backends"

The configure option "--enable-trace-backend" isn't accepted anymore
and we should use "--enable-trace-backends" instead which was
introduce in 2014 and allow multiple backends.

"--enable-trace-backends" was introduced by:
    5b808275f3bb ("trace: Multi-backend tracing")
The backward compatible option "--enable-trace-backend" is removed by
    10229ec3b0ff ("configure: remove backwards-compatibility and obsolete options")

As we already use ./configure options that wouldn't be accepted by
older version of QEMU's configure, we will simply use the new spelling
for the option and avoid trying to detect which spelling to use.

We already make use if "--firmwarepath=" which was introduced by
    3d5eecab4a5a ("Add --firmwarepath to configure")
which already include the new spelling for "--enable-trace-backends".

Signed-off-by: Anthony PERARD <anthony.perard@citrix.com>
Reviewed-by: Jason Andryuk <jandryuk@gmail.com>
master commit: e66d450b6e0ffec635639df993ab43ce28b3383f
master date: 2023-01-11 10:45:29 +0100

3 years agox86/S3: Restore Xen's MSR_PAT value on S3 resume
Andrew Cooper [Tue, 7 Feb 2023 15:56:14 +0000 (16:56 +0100)]
x86/S3: Restore Xen's MSR_PAT value on S3 resume

There are two paths in the trampoline, and Xen's PAT needs setting up in both,
not just the boot path.

Fixes: 4304ff420e51 ("x86/S3: Drop {save,restore}_rest_processor_state() completely")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
master commit: 4d975798e11579fdf405b348543061129e01b0fb
master date: 2023-01-10 21:21:30 +0000

3 years agod/changelog: finish 4.17.0+24-g2f8851c37f-1
Hans van Kranenburg [Wed, 1 Feb 2023 20:52:38 +0000 (21:52 +0100)]
d/changelog: finish 4.17.0+24-g2f8851c37f-1

3 years agoci: Update reason why arm64 crossbuild is disabled
Diederik de Haas [Wed, 21 Dec 2022 23:06:50 +0000 (00:06 +0100)]
ci: Update reason why arm64 crossbuild is disabled

The old reason why it was disabled was bug 982406 'mark markdown
Multi-Arch: foreign', but that was recently fixed.

Trying to enable it revealed another reason why it still doesn't work:
$ eatmydata apt-get build-dep ${HOST_ARCH:+--host-architecture ${HOST_ARCH} -Pcross,nocheck} --no-install-recommends -y $aptopts .
...
Some packages could not be installed. This may mean that you have
requested an impossible situation or if you are using the unstable
distribution that some required packages have not yet been created
or been moved out of Incoming.
The following information may help to resolve the situation:
The following packages have unmet dependencies:
 ocaml:arm64 : Depends: gcc:arm64 but it is not installable
               Depends: binutils:arm64 but it is not installable
E: Unable to correct problems, you have held broken packages.

3 years agod/control: Drop markdown B-D for documentation
Diederik de Haas [Fri, 23 Dec 2022 08:40:25 +0000 (09:40 +0100)]
d/control: Drop markdown B-D for documentation

In upstream commit a2783e97fb220347bcf46583867782712a172710 the build
dependency on markdown was dropped and it has not been needed anymore
since Xen 4.12, so drop it in Debian too.

3 years agod/rules: use pkg-info.mk and do Maintainer parsing in d/rules
Maximilian Engelhardt [Thu, 26 Jan 2023 21:06:50 +0000 (22:06 +0100)]
d/rules: use pkg-info.mk and do Maintainer parsing in d/rules

Use DEB_VERSION and DEB_VERSION_UPSTREAM from
/usr/share/dpkg/pkg-info.mk as suggested by lintian. This fixes
'debian-rules-parses-dpkg-parsechangelog' in the lintian output.

Also move parsing of the Maintainer field in debian/control from our
delta queue to debian/rules and use the newly available DEB_VERSION in
the delta queue.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/xen-hypervisor-common.lintian-overrides: ignore false positive
Maximilian Engelhardt [Thu, 26 Jan 2023 20:18:30 +0000 (21:18 +0100)]
d/xen-hypervisor-common.lintian-overrides: ignore false positive

erroneous 'debian-news-entry-has-unknown-version' is emitted by lintian
due to #1021502.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/control: set Rules-Requires-Root: no
Maximilian Engelhardt [Mon, 23 Jan 2023 19:42:07 +0000 (20:42 +0100)]
d/control: set Rules-Requires-Root: no

As suggested by lintian. There are no differences in the built binaries.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agodebian: remove old leftovers from config file handling
Maximilian Engelhardt [Thu, 5 Jan 2023 23:49:43 +0000 (00:49 +0100)]
debian: remove old leftovers from config file handling

ae40dca3211ec35ca235a8a1f34c37e13093ff0d removed the call to the
debian/ucf-remove-fixup script from debian/rules. However the comment
explaining why this call was there was not removed. Additionally the
override_dh_ucf now only calls dh_ucf without doing anything else.

This commit removes the now unused debian/ucf-remove-fixup script, the
leftover comment referring to it and the dh_ucf override which doesn't
do anything but a call of dh_ucf.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/xen-utils-common.xendomains.default: adjust to upstream template
Maximilian Engelhardt [Mon, 30 Jan 2023 20:20:34 +0000 (21:20 +0100)]
d/xen-utils-common.xendomains.default: adjust to upstream template

Xen upstream sets XENDOMAINS_MIGRATE to any empty string be default. Do
the same in our template file.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/xen-utils-common.xendomains.default: remove XENDOMAINS_SYSRQ
Maximilian Engelhardt [Mon, 30 Jan 2023 20:15:57 +0000 (21:15 +0100)]
d/xen-utils-common.xendomains.default: remove XENDOMAINS_SYSRQ

XENDOMAINS_SYSRQ is currently not supported by our init scripts, so don't
mention it in the default config file.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/shuffle-boot-files: Also handle debug files
Hans van Kranenburg [Sat, 14 Jan 2023 23:40:49 +0000 (00:40 +0100)]
d/shuffle-boot-files: Also handle debug files

See the comment in the change for explanation. We do a fixup for file
names in /boot already, but the files in /usr/lib/debug should get the
same treatment!

Closes: #995233
Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agod/shuffle-boot-files: Add a note about d/not-installed
Hans van Kranenburg [Sat, 14 Jan 2023 23:16:52 +0000 (00:16 +0100)]
d/shuffle-boot-files: Add a note about d/not-installed

Add a hint about the fact that this boot/ location is also present in
d/not-installed. This might help someone looking at all of this for the
first time to discover the puzzle pieces that are involved.

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agodebian: split debug files out of xen-hypervisor-V-F and xen-utils-V
Maximilian Engelhardt [Wed, 18 Jan 2023 22:02:07 +0000 (23:02 +0100)]
debian: split debug files out of xen-hypervisor-V-F and xen-utils-V

The debug files have grown in size over time and can no longer be
considered small.  So we now ship them uncompressed in new -dbg
packages.

The files are installed into /usr/lib/debug at the same path as the
binaries they correspond to, as described in the "Best practices for
debug packages" (Section 6.8.9) in the Debian Developer's Reference.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/control: change Depends from lsb-base to sysvinit-utils
Maximilian Engelhardt [Wed, 18 Jan 2023 23:14:25 +0000 (00:14 +0100)]
d/control: change Depends from lsb-base to sysvinit-utils

lsb-base is now a transitional package depending on sysvinit-utils.
Thus, depending on lsb-base now gives the following lintian error:
E: xen-utils-common: depends-on-obsolete-package Depends: lsb-base

Keep lsb-base as an optional dependency to allow backporting to
bullseye.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/rules: 'dh_missing --fail-missing' is default in dh compat 13
Maximilian Engelhardt [Thu, 5 Jan 2023 23:47:48 +0000 (00:47 +0100)]
d/rules: 'dh_missing --fail-missing' is default in dh compat 13

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agodebian: switch to debhelper compat version 13
Maximilian Engelhardt [Thu, 5 Jan 2023 20:01:13 +0000 (21:01 +0100)]
debian: switch to debhelper compat version 13

Thanks to Diederik de Haas for helping with this.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/control: update build dependency to libext2fs-dev
Maximilian Engelhardt [Fri, 6 Jan 2023 00:02:09 +0000 (01:02 +0100)]
d/control: update build dependency to libext2fs-dev

This also works in bullseye, so backporting is easy.

Signed-off-by: Maximilian Engelhardt <maxi@daemonizer.de>
3 years agod/copyright: rewrite from scratch
Hans van Kranenburg [Sat, 14 Jan 2023 22:14:31 +0000 (23:14 +0100)]
d/copyright: rewrite from scratch

The d/copyright file was very old and outdated. Create an up to date one
now, also using the recommended semi-machine-readable format.

The following files in the upstream source tree were used to produce
this information:

  COPYING
  xen/COPYING
  xen/include/public/COPYING
  xen/common/COPYING
  xen/common/README.source
  xen/common/libelf/COPYING
  xen/crypto/README.source
  xen/include/crypto/README.source
  docs/README.source
  m4/README.source
  stubdom/vtpm/COPYING
  stubdom/COPYING
  tools/firmware/vgabios/COPYING
  tools/include/xen.COPYING.in
  tools/libacpi/COPYING
  tools/libs/guest/COPYING
  tools/xenmon/COPYING
  tools/libs/stat/COPYING
  tools/xenstore/COPYING

If license text is not available on a Debian system by default, the
included text was copied from files in the upstream LICENSES/ directory.

The 4.17 stable branch was used for this. When we advance the upstream
code to e.g. 4.18 we can check if there have been changes made to these
files and update the large copyright file.

Signed-off-by: Hans van Kranenburg <hans@knorrie.org>
3 years agoUpdate changelog for new upstream 4.17.0+24-g2f8851c37f
Maximilian Engelhardt [Wed, 1 Feb 2023 19:28:01 +0000 (20:28 +0100)]
Update changelog for new upstream 4.17.0+24-g2f8851c37f

[git-debrebase changelog: new upstream 4.17.0+24-g2f8851c37f]

3 years agoUpdate to upstream 4.17.0+24-g2f8851c37f
Maximilian Engelhardt [Wed, 1 Feb 2023 19:28:01 +0000 (20:28 +0100)]
Update to upstream 4.17.0+24-g2f8851c37f

[git-debrebase anchor: new upstream 4.17.0+24-g2f8851c37f, merge]

3 years agoRevert "tools/xenstore: simplify loop handling connection I/O"
Jason Andryuk [Thu, 26 Jan 2023 10:00:24 +0000 (11:00 +0100)]
Revert "tools/xenstore: simplify loop handling connection I/O"

I'm observing guest kexec trigger xenstored to abort on a double free.

gdb output:
Program received signal SIGABRT, Aborted.
__pthread_kill_implementation (no_tid=0, signo=6, threadid=140645614258112) at ./nptl/pthread_kill.c:44
44    ./nptl/pthread_kill.c: No such file or directory.
(gdb) bt
    at ./nptl/pthread_kill.c:44
    at ./nptl/pthread_kill.c:78
    at ./nptl/pthread_kill.c:89
    at ../sysdeps/posix/raise.c:26
    at talloc.c:119
    ptr=ptr@entry=0x559fae724290) at talloc.c:232
    at xenstored_core.c:2945
(gdb) frame 5
    at talloc.c:119
119            TALLOC_ABORT("Bad talloc magic value - double free");
(gdb) frame 7
    at xenstored_core.c:2945
2945                talloc_increase_ref_count(conn);
(gdb) p conn
$1 = (struct connection *) 0x559fae724290

Looking at a xenstore trace, we have:
IN 0x559fae71f250 20230120 17:40:53 READ (/local/domain/3/image/device-model-dom
id )
wrl: dom    0      1  msec      10000 credit     1000000 reserve        100 disc
ard
wrl: dom    3      1  msec      10000 credit     1000000 reserve        100 disc
ard
wrl: dom    0      0  msec      10000 credit     1000000 reserve          0 disc
ard
wrl: dom    3      0  msec      10000 credit     1000000 reserve          0 disc
ard
OUT 0x559fae71f250 20230120 17:40:53 ERROR (ENOENT )
wrl: dom    0      1  msec      10000 credit     1000000 reserve        100 disc
ard
wrl: dom    3      1  msec      10000 credit     1000000 reserve        100 disc
ard
IN 0x559fae71f250 20230120 17:40:53 RELEASE (3 )
DESTROY watch 0x559fae73f630
DESTROY watch 0x559fae75ddf0
DESTROY watch 0x559fae75ec30
DESTROY watch 0x559fae75ea60
DESTROY watch 0x559fae732c00
DESTROY watch 0x559fae72cea0
DESTROY watch 0x559fae728fc0
DESTROY watch 0x559fae729570
DESTROY connection 0x559fae724290
orphaned node /local/domain/3/device/suspend/event-channel deleted
orphaned node /local/domain/3/device/vbd/51712 deleted
orphaned node /local/domain/3/device/vkbd/0 deleted
orphaned node /local/domain/3/device/vif/0 deleted
orphaned node /local/domain/3/control/shutdown deleted
orphaned node /local/domain/3/control/feature-poweroff deleted
orphaned node /local/domain/3/control/feature-reboot deleted
orphaned node /local/domain/3/control/feature-suspend deleted
orphaned node /local/domain/3/control/feature-s3 deleted
orphaned node /local/domain/3/control/feature-s4 deleted
orphaned node /local/domain/3/control/sysrq deleted
orphaned node /local/domain/3/data deleted
orphaned node /local/domain/3/drivers deleted
orphaned node /local/domain/3/feature deleted
orphaned node /local/domain/3/attr deleted
orphaned node /local/domain/3/error deleted
orphaned node /local/domain/3/console/backend-id deleted

and no further output.

The trace shows that DESTROY was called for connection 0x559fae724290,
but that is the same pointer (conn) main() was looping through from
connections.  So it wasn't actually removed from the connections list?

Reverting commit e8e6e42279a5 "tools/xenstore: simplify loop handling
connection I/O" fixes the abort/double free.  I think the use of
list_for_each_entry_safe is incorrect.  list_for_each_entry_safe makes
traversal safe for deleting the current iterator, but RELEASE/do_release
will delete some other entry in the connections list.  I think the
observed abort is because list_for_each_entry has next pointing to the
deleted connection, and it is used in the subsequent iteration.

Add a comment explaining the unsuitability of list_for_each_entry_safe.
Also notice that the old code takes a reference on next which would
prevents a use-after-free.

This reverts commit e8e6e42279a5723239c5c40ba4c7f579a979465d.

This is XSA-425/CVE-2022-42330.

Fixes: e8e6e42279a5 ("tools/xenstore: simplify loop handling connection I/O")
Signed-off-by: Jason Andryuk <jandryuk@gmail.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
Reviewed-by: Julien Grall <jgrall@amazon.com>
3 years agodebian/changelog: finish 4.17.0-1
Maximilian Engelhardt [Wed, 21 Dec 2022 21:36:11 +0000 (22:36 +0100)]
debian/changelog: finish 4.17.0-1

3 years agod/control: update Build-Depends for ocaml
Maximilian Engelhardt [Wed, 21 Dec 2022 21:19:39 +0000 (22:19 +0100)]
d/control: update Build-Depends for ocaml

ocaml-native-compilers is not shipped in Debian since buster, ocaml-nox
is a transitional package for ocaml in unstable/testing.
Since ocaml depends on ocaml-nox in bullseye, it doesn't affect
backports.

3 years agod/control: Update Standards-Version to 4.6.2
Maximilian Engelhardt [Wed, 21 Dec 2022 21:18:30 +0000 (22:18 +0100)]
d/control: Update Standards-Version to 4.6.2

no changes needed

3 years agoUpdate changelog for new upstream 4.17.0
Maximilian Engelhardt [Wed, 21 Dec 2022 21:03:44 +0000 (22:03 +0100)]
Update changelog for new upstream 4.17.0

[git-debrebase changelog: new upstream 4.17.0]

3 years agoUpdate to upstream 4.17.0
Maximilian Engelhardt [Wed, 21 Dec 2022 21:03:43 +0000 (22:03 +0100)]
Update to upstream 4.17.0

[git-debrebase anchor: new upstream 4.17.0, merge]

3 years agotools/oxenstored: Render backtraces more nicely in Syslog
Andrew Cooper [Thu, 1 Dec 2022 21:06:25 +0000 (21:06 +0000)]
tools/oxenstored: Render backtraces more nicely in Syslog

fallback_exception_handler feeds a string with embedded newlines directly into
syslog().  While this is an improvement on getting nothing, syslogd escapes
all control characters it gets, and emits one (long) log line.

Fix the problem generally in the syslog stub.  As we already have a local copy
of the string, split it in place and emit one syslog() call per line.

Also tweak Logging.msg_of to avoid putting an extra newline on a string which
already ends with one.

Fixes: ee7815f49faf ("tools/oxenstored: Set uncaught exception handler")
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit d2162d884cba0ff7b2ac0d832f4e044444bda2e1)

3 years agotools/oxenstored/syslog: Avoid potential NULL dereference
Edwin Török [Tue, 8 Nov 2022 14:24:19 +0000 (14:24 +0000)]
tools/oxenstored/syslog: Avoid potential NULL dereference

strdup() may return NULL.  Check for this before passing to syslog().

Drop const from c_msg.  It is bogus, as demonstrated by the need to cast to
void * in order to free the memory.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit acd3fb6d65905f8a185dcb9fe6a330a591b96203)

3 years agotools/oxenstored: Set uncaught exception handler
Edwin Török [Mon, 7 Nov 2022 17:41:36 +0000 (17:41 +0000)]
tools/oxenstored: Set uncaught exception handler

Unhandled exceptions go to stderr by default, but this doesn't typically work
for oxenstored because:
 * daemonize reopens stderr as /dev/null
 * systemd redirects stderr to /dev/null too

Debugging an unhandled exception requires reproducing the issue locally when
using --no-fork, and is not conducive to figuring out what went wrong on a
remote system.

Install a custom handler which also tries to render the backtrace to the
configured syslog facility, and DAEMON|ERR otherwise.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit ee7815f49faf743e960dac9e72809eb66393bc6d)

3 years agotools/oxenstored: Log live update issues at warning level
Edwin Török [Tue, 8 Nov 2022 08:57:47 +0000 (08:57 +0000)]
tools/oxenstored: Log live update issues at warning level

During live update, oxenstored tries a best effort approach to recover as many
domains and information as possible even if it encounters errors restoring
some domains.

However, logging about misunderstood input is more severe than simply info.
Log it at warning instead.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 3f02e0a70fe9f8143454b742563433958d4a87f8)

3 years agotools/oxenstored: Keep /dev/xen/evtchn open across live update
Edwin Török [Thu, 3 Nov 2022 15:31:39 +0000 (15:31 +0000)]
tools/oxenstored: Keep /dev/xen/evtchn open across live update

Closing the evtchn handle will unbind and free all local ports.  The new
xenstored would need to rebind all evtchns, which is work that we don't want
or need to be doing during the critical handover period.

However, it turns out that the Windows PV drivers also rebind their local port
too across suspend/resume, leaving (o)xenstored with a stale idea of the
remote port to use.  In this case, reusing the established connection is the
only robust option.

Therefore:
 * Have oxenstored open /dev/xen/evtchn without CLOEXEC at start of day.
 * Extend the handover information with the evtchn fd, domexc virq local port,
   and the local port number for each domain connection.
 * Have (the new) oxenstored recover the open handle using Xeneventchn.fdopen,
   and use the provided local ports rather than trying to rebind them.

When this new information isn't present (i.e. live updating from an oxenstored
prior to this change), the best-effort status quo will have to do.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 9b224c25293a53fcbe32da68052d861dda71a6f4)

3 years agotools/oxenstored: Rework Domain evtchn handling to use port_pair
Andrew Cooper [Wed, 30 Nov 2022 11:59:34 +0000 (11:59 +0000)]
tools/oxenstored: Rework Domain evtchn handling to use port_pair

Inter-domain event channels are always a pair of local and remote ports.
Right now the handling is asymmetric, caused by the fact that the evtchn is
bound after the associated Domain object is constructed.

First, move binding of the event channel into the Domain.make() constructor.
This means the local port no longer needs to be an option.  It also removes
the final callers of Domain.bind_interdomain.

Next, introduce a new port_pair type to encapsulate the fact that these two
should be updated together, and replace the previous port and remote_port
fields.  This refactoring also changes the Domain.get_port interface (removing
an option) so take the opportunity to name it get_local_port instead.

Also, this fixes a use-after-free risk with Domain.close.  Once the evtchn has
been unbound, the same local port number can be reused for a different
purpose, so explicitly invalidate the ports to prevent their accidental misuse
in the future.

This also cleans up some of the debugging, to always print a port pair.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit df2db174b36eba67c218763ef621c67912202fc6)

3 years agotools/oxenstored: Implement Domain.rebind_evtchn
Andrew Cooper [Wed, 30 Nov 2022 11:55:58 +0000 (11:55 +0000)]
tools/oxenstored: Implement Domain.rebind_evtchn

Generally speaking, the event channel local/remote port is fixed for the
lifetime of the associated domain object.  The exception to this is a
secondary XS_INTRODUCE (defined to re-bind to a new event channel) which pokes
around at the domain object's internal state.

We need to refactor the evtchn handling to support live update, so start by
moving the relevant manipulation into Domain.

No practical change.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit aecdc28d9538ca2a1028ef9bc6550cb171dbbed4)

3 years agotools/oxenstored: Rename some 'port' variables to 'remote_port'
Andrew Cooper [Wed, 30 Nov 2022 03:17:28 +0000 (03:17 +0000)]
tools/oxenstored: Rename some 'port' variables to 'remote_port'

This will make the logic clearer when we plumb local_port through these
functions.

While doing this, rearrange the construct in Domains.create0 to separate the
remote port handling from the interface handling.  (The interface logic is
dubious in several ways, but not altered by this cleanup.)

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 31fbee749a75621039ca601eaee7222050a7dd83)

3 years agotools/oxenstored: Bind the DOM_EXC VIRQ in in Event.init()
Andrew Cooper [Tue, 29 Nov 2022 21:05:43 +0000 (21:05 +0000)]
tools/oxenstored: Bind the DOM_EXC VIRQ in in Event.init()

Xenstored always needs to bind the DOM_EXC VIRQ.

Instead of doing it shortly after the call to Event.init(), do it in the
constructor directly.  This removes the need for the field to be a mutable
option.

It will also simplify a future change to support live update.  Rename the
field from virq_port (which could be any VIRQ) to it's proper name.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 9804a5db435fe40c8ded8cf36c2d2b2281c56f1d)

3 years agotools/oxenstored: Style fixes to Domain
Andrew Cooper [Wed, 30 Nov 2022 14:56:43 +0000 (14:56 +0000)]
tools/oxenstored: Style fixes to Domain

This file has some style problems so severe that they interfere with the
readability of the subsequent bugfix patches.

Fix these issues ahead of time, to make the subsequent changes more readable.

No functional change.

Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Edwin Török <edvin.torok@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit b45bfaf359e4821b1bf98a4fcd194d7fd176f167)

3 years agotools/ocaml/evtchn: Extend the init() binding with a cloexec flag
Edwin Török [Thu, 3 Nov 2022 14:50:38 +0000 (14:50 +0000)]
tools/ocaml/evtchn: Extend the init() binding with a cloexec flag

For live update, oxenstored wants to clear CLOEXEC on the evtchn handle, so it
survives the execve() into the new oxenstored.

Have the new interface match how cloexec works in other Ocaml standard
libraries.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 9bafe4a53306e7aa2ce6ffc96f7477c6f329f7a7)

3 years agotools/ocaml/evtchn: Add binding for xenevtchn_fdopen()
Edwin Török [Mon, 14 Nov 2022 13:36:19 +0000 (13:36 +0000)]
tools/ocaml/evtchn: Add binding for xenevtchn_fdopen()

For live update, the new oxenstored needs to reconstruct an evtchn object
around an existing file descriptor.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 7ba68a6c558e1fd811c95cb7215a5cd07a3cc2ea)

3 years agotools/ocaml/evtchn: OCaml 5 support, fix potential resource leak
Edwin Török [Tue, 18 Jan 2022 15:04:48 +0000 (15:04 +0000)]
tools/ocaml/evtchn: OCaml 5 support, fix potential resource leak

There is no binding for xenevtchn_close().  In principle, this is a resource
leak, but the typical usage is as a singleton that lives for the lifetime of
the program.

Ocaml 5 no longer permits storing a naked C pointer in an Ocaml value.

Therefore, use a Custom block.  This allows us to use the finaliser callback
to call xenevtchn_close(), if the Ocaml object goes out of scope.

Signed-off-by: Edwin Török <edvin.torok@citrix.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit 22d5affdf0cecfa6faae46fbaec68b8018835220)

3 years agotools/oxenstored: Fix incorrect scope after an if statement
Andrew Cooper [Fri, 11 Nov 2022 18:50:34 +0000 (18:50 +0000)]
tools/oxenstored: Fix incorrect scope after an if statement

A debug statement got inserted into a single-expression if statement.

Insert brackets to give the intended meaning, rather than the actual meaning
where the "let con = Connections..." is outside and executed unconditionally.

This results in some unnecessary ring checks for domains which otherwise have
IO credit.

Fixes: 42f0581a91d4 ("tools/oxenstored: Implement live update for socket connections")
Reported-by: Edwin Török <edvin.torok@citrix.com>
Signed-off-by: Andrew Cooper <andrew.cooper3@citrix.com>
Acked-by: Christian Lindig <christian.lindig@citrix.com>
(cherry picked from commit ee36179371fd4215a43fb179be2165f65c1cd1cd)

3 years agox86/time: prevent overflow with high frequency TSCs
Neowutran [Tue, 20 Dec 2022 12:46:38 +0000 (13:46 +0100)]
x86/time: prevent overflow with high frequency TSCs

Make sure tsc_khz is promoted to a 64-bit type before multiplying by
1000 to avoid an 'overflow before widen' bug. Otherwise just above
4.294GHz the value will overflow. Processors with clocks this high are
now in production and require this to work correctly.

Signed-off-by: Neowutran <xen@neowutran.ovh>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
master commit: ad15a0a8ca2515d8ac58edfc0bc1d3719219cb77
master date: 2022-12-19 11:34:16 +0100

3 years agoEFI: relocate the ESRT when booting via multiboot2
Demi Marie Obenour [Tue, 20 Dec 2022 12:46:09 +0000 (13:46 +0100)]
EFI: relocate the ESRT when booting via multiboot2

This was missed in the initial patchset.

Move efi_relocate_esrt() up to avoid adding a forward declaration.

Signed-off-by: Demi Marie Obenour <demi@invisiblethingslab.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
master commit: 8d7acf3f7d8d2555c78421dced45bc49f79ae806
master date: 2022-12-14 12:00:35 +0100

3 years agoioreq_broadcast(): accept partial broadcast success
Per Bilse [Tue, 20 Dec 2022 12:45:38 +0000 (13:45 +0100)]
ioreq_broadcast(): accept partial broadcast success

Avoid incorrectly triggering an error when a broadcast buffered ioreq
is not handled by all registered clients, as long as the failure is
strictly because the client doesn't handle buffered ioreqs.

Signed-off-by: Per Bilse <per.bilse@citrix.com>
Reviewed-by: Paul Durrant <paul@xen.org>
master commit: a44734df6c24fadbdb001f051cc5580c467caf7d
master date: 2022-12-07 12:17:30 +0100

3 years agox86/HVM: don't mark evtchn upcall vector as pending when vLAPIC is disabled
Jan Beulich [Tue, 20 Dec 2022 12:45:07 +0000 (13:45 +0100)]
x86/HVM: don't mark evtchn upcall vector as pending when vLAPIC is disabled

Linux'es relatively new use of HVMOP_set_evtchn_upcall_vector has
exposed a problem with the marking of the respective vector as
pending: For quite some time Linux has been checking whether any stale
ISR or IRR bits would still be set while preparing the LAPIC for use.
This check is now triggering on the upcall vector, as the registration,
at least for APs, happens before the LAPIC is actually enabled.

In software-disabled state an LAPIC would not accept any interrupt
requests and hence no IRR bit would newly become set while in this
state. As a result it is also wrong for us to mark the upcall vector as
having a pending request when the vLAPIC is in this state.

To compensate for the "enabled" check added to the assertion logic, add
logic to (conditionally) mark the upcall vector as having a request
pending at the time the LAPIC is being software-enabled by the guest.
Note however that, like for the pt_may_unmask_irq() we already have
there, long term we may need to find a different solution. This will be
especially relevant in case yet better LAPIC acceleration would
eliminate notifications of guest writes to this and other registers.

Fixes: 7b5b8ca7dffd ("x86/upcall: inject a spurious event after setting upcall vector")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Reviewed-by: Juergen Gross <jgross@suse.com>
master commit: f5d0279839b58cb622f0995dbf9cff056f03082e
master date: 2022-12-06 13:51:49 +0100

3 years agox86/Viridian: don't mark IRQ vectors as pending when vLAPIC is disabled
Jan Beulich [Tue, 20 Dec 2022 12:44:38 +0000 (13:44 +0100)]
x86/Viridian: don't mark IRQ vectors as pending when vLAPIC is disabled

In software-disabled state an LAPIC does not accept any interrupt
requests and hence no IRR bit would newly become set while in this
state. As a result it is also wrong for us to mark Viridian IPI or timer
vectors as having a pending request when the vLAPIC is in this state.
Such interrupts are simply lost.

Introduce a local variable in send_ipi() to help readability.

Fixes: fda96b7382ea ("viridian: add implementation of the HvSendSyntheticClusterIpi hypercall")
Fixes: 26fba3c85571 ("viridian: add implementation of synthetic timers")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
Reviewed-by: Paul Durrant <paul@xen.org>
master commit: 831419f82913417dee4e5b0f80769c5db590540b
master date: 2022-12-02 10:35:32 +0100

3 years agox86/HVM: don't mark external IRQs as pending when vLAPIC is disabled
Jan Beulich [Tue, 20 Dec 2022 12:44:07 +0000 (13:44 +0100)]
x86/HVM: don't mark external IRQs as pending when vLAPIC is disabled

In software-disabled state an LAPIC does not accept any interrupt
requests and hence no IRR bit would newly become set while in this
state. As a result it is also wrong for us to mark IO-APIC or MSI
originating vectors as having a pending request when the vLAPIC is in
this state. Such interrupts are simply lost.

Introduce (IO-APIC) or re-use (MSI) a local variable to help
readability.

Fixes: 4fe21ad3712e ("This patch add virtual IOAPIC support for VMX guest")
Fixes: 85715f4bc7c9 ("MSI 5/6: add MSI support to passthrough HVM domain")
Signed-off-by: Jan Beulich <jbeulich@suse.com>
Acked-by: Andrew Cooper <andrew.cooper3@citrix.com>
master commit: f1d7aac1e3c3cd164e17d41791a575a5c3e87121
master date: 2022-12-02 10:35:01 +0100

3 years agox86/pvh: do not forward MADT Local APIC NMI structures to dom0
Roger Pau Monné [Tue, 20 Dec 2022 12:43:37 +0000 (13:43 +0100)]
x86/pvh: do not forward MADT Local APIC NMI structures to dom0

Currently Xen will passthrough any Local APIC NMI Structure found in
the native ACPI MADT table to a PVH dom0.  This is wrong because PVH
doesn't have access to the physical local APIC, and instead gets an
emulated local APIC by Xen, that doesn't have the LINT0 or LINT1
pins wired to anything.  Furthermore the ACPI Processor UIDs used in
the APIC NMI Structures are likely to not match the ones generated by
Xen for the Local x2APIC Structures, creating confusion to dom0.

Fix this by removing the logic to passthrough the Local APIC NMI
Structure for PVH dom0.

Fixes: 1d74282c45 ('x86: setup PVHv2 Dom0 ACPI tables')
Signed-off-by: Roger Pau Monné <roger.pau@citrix.com>
Reviewed-by: Jan Beulich <jbeulich@suse.com>
master commit: b39e6385250ccef9509af0eab9003ad5c1478842
master date: 2022-12-02 10:33:40 +0100