The ``loaded=on`` option in the command line or QMP ``object-add`` either had
no effect (if ``loaded`` was the last option) or caused options to be
effectively ignored as if they were not given. The property is therefore
useless and was deprecated in 6.0; make it read-only now.
The patch is best reviewed with "-b".
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Code consuming the "crypto/tlscreds*.h" APIs doesn't need
to access its internals. Move the structure definitions to
the "tlscredspriv.h" private header (only accessible by
implementations). The public headers (in include/) still
forward-declare the structures typedef.
Note, tlscreds.c and 3 of the 5 modified source files already
include "tlscredspriv.h", so only add it to tls-cipher-suites.c
and tlssession.c.
Removing the internals from the public header solves a bug
introduced by commit 7de2e85653 ("yank: Unregister function
when using TLS migration") which made migration/qemu-file-channel.c
include "io/channel-tls.h", itself sometime depends on GNUTLS,
leading to a build failure on OSX:
[2/35] Compiling C object libmigration.fa.p/migration_qemu-file-channel.c.o
FAILED: libmigration.fa.p/migration_qemu-file-channel.c.o
cc -Ilibmigration.fa.p -I. -I.. -Iqapi [ ... ] -o libmigration.fa.p/migration_qemu-file-channel.c.o -c ../migration/qemu-file-channel.c
In file included from ../migration/qemu-file-channel.c:29:
In file included from include/io/channel-tls.h:26:
In file included from include/crypto/tlssession.h:24:
include/crypto/tlscreds.h:28:10: fatal error: 'gnutls/gnutls.h' file not found
#include <gnutls/gnutls.h>
^~~~~~~~~~~~~~~~~
1 error generated.
Reported-by: Stefan Weil <sw@weilnetz.de>
Suggested-by: Daniel P. Berrangé <berrange@redhat.com>
Resolves: https://gitlab.com/qemu-project/qemu/-/issues/407
Fixes: 7de2e85653 ("yank: Unregister function when using TLS migration")
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The condition being tested has never been set since the day the code was
first introduced.
Reviewed-by: Willian Rampazzo <willianr@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210514120415.1368922-8-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
This patch adds reload interface for QCryptoTLSCredsClass and implements
the interface for QCryptoTLSCredsX509.
Signed-off-by: Zihao Chang <changzihao1@huawei.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20210316075845.1476-2-changzihao1@huawei.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
If you set the loaded property to true when it was already true, the
state is overwritten without freeing the old state first. Change the
set_loaded callback so that it always frees the old state (which is a
no-op if nothing was loaded) and only then load if requestsd.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
This patch fixes error style problems found by checkpatch.pl:
ERROR: spaces required around that '*'
ERROR: space required after that ','
ERROR: spaces required around that '|'
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Liyang Shi <shiliyang@huawei.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
The object_property_set_FOO() setters take property name and value in
an unusual order:
void object_property_set_FOO(Object *obj, FOO_TYPE value,
const char *name, Error **errp)
Having to pass value before name feels grating. Swap them.
Same for object_property_set(), object_property_get(), and
object_property_parse().
Convert callers with this Coccinelle script:
@@
identifier fun = {
object_property_get, object_property_parse, object_property_set_str,
object_property_set_link, object_property_set_bool,
object_property_set_int, object_property_set_uint, object_property_set,
object_property_set_qobject
};
expression obj, v, name, errp;
@@
- fun(obj, v, name, errp)
+ fun(obj, name, v, errp)
Chokes on hw/arm/musicpal.c's lcd_refresh() with the unhelpful error
message "no position information". Convert that one manually.
Fails to convert hw/arm/armsse.c, because Coccinelle gets confused by
ARMSSE being used both as typedef and function-like macro there.
Convert manually.
Fails to convert hw/rx/rx-gdbsim.c, because Coccinelle gets confused
by RXCPU being used both as typedef and function-like macro there.
Convert manually. The other files using RXCPU that way don't need
conversion.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-Id: <20200707160613.848843-27-armbru@redhat.com>
[Straightforwad conflict with commit 2336172d9b "audio: set default
value for pcspk.iobase property" resolved]
The only way object_property_add() can fail is when a property with
the same name already exists. Since our property names are all
hardcoded, failure is a programming error, and the appropriate way to
handle it is passing &error_abort.
Same for its variants, except for object_property_add_child(), which
additionally fails when the child already has a parent. Parentage is
also under program control, so this is a programming error, too.
We have a bit over 500 callers. Almost half of them pass
&error_abort, slightly fewer ignore errors, one test case handles
errors, and the remaining few callers pass them to their own callers.
The previous few commits demonstrated once again that ignoring
programming errors is a bad idea.
Of the few ones that pass on errors, several violate the Error API.
The Error ** argument must be NULL, &error_abort, &error_fatal, or a
pointer to a variable containing NULL. Passing an argument of the
latter kind twice without clearing it in between is wrong: if the
first call sets an error, it no longer points to NULL for the second
call. ich9_pm_add_properties(), sparc32_ledma_realize(),
sparc32_dma_realize(), xilinx_axidma_realize(), xilinx_enet_realize()
are wrong that way.
When the one appropriate choice of argument is &error_abort, letting
users pick the argument is a bad idea.
Drop parameter @errp and assert the preconditions instead.
There's one exception to "duplicate property name is a programming
error": the way object_property_add() implements the magic (and
undocumented) "automatic arrayification". Don't drop @errp there.
Instead, rename object_property_add() to object_property_try_add(),
and add the obvious wrapper object_property_add().
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200505152926.18877-15-armbru@redhat.com>
[Two semantic rebase conflicts resolved]
qcrypto_tls_creds_load_cert() passes uninitialized GError *gerr by
reference to g_file_get_contents(). When g_file_get_contents() fails,
it'll try to set a GError. Unless @gerr is null by dumb luck, this
logs a ERROR_OVERWRITTEN_WARNING warning message and leaves @gerr
unchanged. qcrypto_tls_creds_load_cert() then dereferences the
uninitialized @gerr.
Fix by initializing @gerr properly.
Fixes: 9a2fd4347c
Cc: "Daniel P. Berrangé" <berrange@redhat.com>
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20191204093625.14836-2-armbru@redhat.com>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Simplify cleanup paths by using glib's auto cleanup macros for stack
variables, allowing several goto jumps / labels to be eliminated.
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
It's either "GNU *Library* General Public License version 2" or "GNU
Lesser General Public License version *2.1*", but there was no "version
2.0" of the "Lesser" license. So assume that version 2.1 is meant here.
Signed-off-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
gnutls 3.0.0 was released in 2011 and all the distros that are build
target platforms for QEMU [1] include it:
RHEL-7: 3.1.18
Debian (Stretch): 3.5.8
Debian (Jessie): 3.3.8
OpenBSD (ports): 3.5.18
FreeBSD (ports): 3.5.18
OpenSUSE Leap 15: 3.6.2
Ubuntu (Xenial): 3.4.10
macOS (Homebrew): 3.5.19
Based on this, it is reasonable to require gnutls >= 3.1.18 in QEMU
which allows for all conditional version checks in the code to be
removed.
[1] https://qemu.weilnetz.de/doc/qemu-doc.html#Supported-build-platforms
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
When pulling in headers that are in the same directory as the C file (as
opposed to one in include/), we should use its relative path, without a
directory.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Tested-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Acked-by: Daniel P. Berrangé <berrange@redhat.com>
gnutls commit 846753877d renamed LIBGNUTLS_VERSION_NUMBER to GNUTLS_VERSION_NUMBER.
If using gnutls before that verion, we'll get the below warning:
crypto/tlscredsx509.c:618:5: warning: "GNUTLS_VERSION_NUMBER" is not defined
Because gnutls 3.x still defines LIBGNUTLS_VERSION_NUMBER for back compat, Let's
use LIBGNUTLS_VERSION_NUMBER instead of GNUTLS_VERSION_NUMBER to fix building
complaint.
Signed-off-by: Gonglei <arei.gonglei@huawei.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Give slightly more information when certification loading fails.
Rather than have no information, you now get gnutls's only slightly
less unhelpful error messages.
Signed-off-by: Alex Bligh <alex@alex.org.uk>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Commit 57cb38b included qapi/error.h into qemu/osdep.h to get the
Error typedef. Since then, we've moved to include qemu/osdep.h
everywhere. Its file comment explains: "To avoid getting into
possible circular include dependencies, this file should not include
any other QEMU headers, with the exceptions of config-host.h,
compiler.h, os-posix.h and os-win32.h, all of which are doing a
similar job to this file and are under similar constraints."
qapi/error.h doesn't do a similar job, and it doesn't adhere to
similar constraints: it includes qapi-types.h. That's in excess of
100KiB of crap most .c files don't actually need.
Add the typedef to qemu/typedefs.h, and include that instead of
qapi/error.h. Include qapi/error.h in .c files that need it and don't
get it now. Include qapi-types.h in qom/object.h for uint16List.
Update scripts/clean-includes accordingly. Update it further to match
reality: replace config.h by config-target.h, add sysemu/os-posix.h,
sysemu/os-win32.h. Update the list of includes in the qemu/osdep.h
comment quoted above similarly.
This reduces the number of objects depending on qapi/error.h from "all
of them" to less than a third. Unfortunately, the number depending on
qapi-types.h shrinks only a little. More work is needed for that one.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
[Fix compilation without the spice devel packages. - Paolo]
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This converts the tlscredsx509, tlscredsanon and secret objects
to register their properties against the class rather than object.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Clean up includes so that osdep.h is included first and headers
which it implies are not included manually.
This commit was created with scripts/clean-includes.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-id: 1453832250-766-3-git-send-email-peter.maydell@linaro.org
Make use of the QCryptoSecret object to support loading of
encrypted x509 keys. The optional 'passwordid' parameter
to the tls-creds-x509 object type, provides the ID of a
secret object instance that holds the decryption password
for the PEM file.
# printf "123456" > mypasswd.txt
# $QEMU \
-object secret,id=sec0,filename=mypasswd.txt \
-object tls-creds-x509,passwordid=sec0,id=creds0,\
dir=/home/berrange/.pki/qemu,endpoint=server \
-vnc :1,tls-creds=creds0
This requires QEMU to be linked to GNUTLS >= 3.1.11. If
GNUTLS is too old an error will be reported if an attempt
is made to pass a decryption password.
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
In qcrypto_tls_creds_get_path() coverity complains that
we are checking '*creds' for NULL, despite having
dereferenced it previously. This is harmless bug due
to fact that the trace call was too early. Moving it
after the cleanup gets the desired semantics.
In qcrypto_tls_creds_check_cert_key_purpose() coverity
complains that we're passing a pointer to a previously
free'd buffer into gnutls_x509_crt_get_key_purpose_oid()
This is harmless because we're passing a size == 0, so
gnutls won't access the buffer, but rather just report
what size it needs to be. We can avoid it though by
explicitly setting the buffer to NULL after free'ing
it.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Michael Tokarev <mjt@tls.msk.ru>
The qcrypto_tls_creds_x509_sanity_check() checks whether
certs exist by calling access(). It is valid for this
method to be invoked with certfile==NULL though, since
for client credentials the cert is optional. This caused
it to call access(NULL), which happens to be harmless on
current Linux, but should none the less be avoided.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
The QCryptoTLSCredsX509 object was not free'ing the allocated
gnutls_dh_params_t data when unloading the credentials
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
If the administrator incorrectly sets up their x509 certificates,
the errors seen at runtime during connection attempts are very
obscure and difficult to diagnose. This has been a particular
problem for people using openssl to generate their certificates
instead of the gnutls certtool, because the openssl tools don't
turn on the various x509 extensions that gnutls expects to be
present by default.
This change thus adds support in the TLS credentials object to
sanity check the certificates when QEMU first loads them. This
gives the administrator immediate feedback for the majority of
common configuration mistakes, reducing the pain involved in
setting up TLS. The code is derived from equivalent code that
has been part of libvirt's TLS support and has been seen to be
valuable in assisting admins.
It is possible to disable the sanity checking, however, via
the new 'sanity-check' property on the tls-creds object type,
with a value of 'no'.
Unit tests are included in this change to verify the correctness
of the sanity checking code in all the key scenarios it is
intended to cope with. As part of the test suite, the pkix_asn1_tab.c
from gnutls is imported. This file is intentionally copied from the
(long since obsolete) gnutls 1.6.3 source tree, since that version
was still under GPLv2+, rather than the GPLv3+ of gnutls >= 2.0.
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Introduce a QCryptoTLSCredsX509 class which is used to
manage x509 certificate TLS credentials. This will be
the preferred credential type offering strong security
characteristics
Example CLI configuration:
$QEMU -object tls-creds-x509,id=tls0,endpoint=server,\
dir=/path/to/creds/dir,verify-peer=yes
The 'id' value in the -object args will be used to associate the
credentials with the network services. For example, when the VNC
server is later converted it would use
$QEMU -object tls-creds-x509,id=tls0,.... \
-vnc 127.0.0.1:1,tls-creds=tls0
Signed-off-by: Daniel P. Berrange <berrange@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Paolo Bonzini
Philippe Mathieu-Daudé
Daniel P. Berrangé
Zihao Chang
Kevin Wolf
shiliyang
Markus Armbruster
Thomas Huth
Michael S. Tsirkin
Gonglei
Alex Bligh
Peter Maydell