It is a bit of a pain to be forced to run configure before being able
to use cscope and friends. Add back the rules to build them in-tree
as before commit a5665051.

Fixes: a5665051 ("configure: integrate Meson in the build system")
Signed-off-by: Greg Kurz <groug@kaod.org>
Message-Id: <159897001005.442705.16516671603870288336.stgit@bahia.lan>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

This will be useful in the future to generate configure
command line parsing from meson_options.txt.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

The missing "/all" suffix prevents the pc-bios/ parts of the build
from running.

In the meanwhile, -Wall has moved from QEMU_CFLAGS to CFLAGS.  Simplify
everything by not passing down CFLAGS, and add -Wall in the recursive
Makefiles.

Reported-by: Miroslav Rezanina <mrezanin@redhat.com>
Reviewed-by: Miroslav Rezanina <mrezanin@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Fixes: 5e6d1573 ("remove Makefile.target", 2020-08-21)
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

We do not need to ask cmake for the dependencies, so just use the
pkg-config mechanism.  Keep "auto" for SDL so that it tries using
sdl-config too.

The documentation is adjusted to use SDL2_image as the example,
rather than SDL which does not use the "pkg-config" method.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Left out in commit 22a87800 ("configure: expand path variables for
meson configure", 2020-08-21), do it now.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Pixman used to be included directly in QEMU_CFLAGS and therefore the
include path was added to every compiler invocation.  Now that (just
like basically everything else) it is a separate dependency, we
need to add it to all build target, especially UI modules that need
it due to their including ui/console.h.

Reported-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Stefan Weil <sw@weilnetz.de>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200830204640.482214-1-sw@weilnetz.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20200826110419.528931-9-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

When cross-compiling, by default qemu_mandir is 'c:\Program
Files\QEMU', which is not recognized as being an absolute path, and
meson will end up adding the prefix again.

Use the pre-prefixed meson mandir option instead.

Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-8-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

When cross-compiling, by default qemu_docdir is 'c:\Program Files\QEMU\'
which is not recognized as being an absolute path, and meson will end up
adding the prefix again.

Add an option to pass docdir location to meson, pre-prefixed like we do
with other directories, build qemu_docdir with the common suffix and use
that instead of config_host['qemu_docdir'].

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-7-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

When cross-compiling, by default qemu_datadir is 'c:\Program
Files\QEMU', which is not recognized as being an absolute path, and
meson will end up adding the prefix again.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-6-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

The following patches will make use of it to fix installation paths.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-5-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

As documented in --help for --docdir.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-4-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Otherwise, we may accept very strange directory names...

While at it, quote the variables.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-3-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

The value is used to construct conf/mod/data directories.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826110419.528931-2-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Meson includes the same logic that tries to look for -lz if
pkg-config (and cmake) cannot find zlib.  The undocumented
--disable-zlib-test option becomes a no-op.

There is still an instance of "-lz" in the LIBS directory.
It will go away as soon as tests are converted to meson,
because the zlib dependency does not propagate from libblock.fa
to the Makefile-build unit tests.

Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Now covered by meson

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826130622.553318-7-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826130622.553318-6-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826130622.553318-5-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826130622.553318-4-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826130622.553318-3-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20200826130622.553318-2-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

Configure used to probe for SDL2_image in sdl_image_probe (). Meson
should do the same.

This fixes the following error on my system:

Run-time dependency sdl2 found: YES 2.0.8
Found CMake: /usr/bin/cmake (3.17.0)
Run-time dependency sdl-image found: NO (tried pkgconfig and cmake)

../qemu-master/meson.build:256:2: ERROR: Dependency "sdl-image" not
found, tried pkgconfig and cmake

A full log can be found at /home/ruemelin/rpmbuild/BUILD
/qemu-5.1.50-build/meson-logs/meson-log.txt

ERROR: meson setup failed

Signed-off-by: Volker Rümelin <vr_qemu@t-online.de>
Message-Id: <20200829104158.7461-1-vr_qemu@t-online.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

The qcow2 driver needs the zlib dependency.  While emulators
provided it through the migration code, this is not true of
the tools.  Move the dependency from the qcow1 rule directly
into block_ss so that it is included unconditionally.

Fixes build with --disable-qcow1.

Reported-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Cc: qemu-block@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

This version includes an important bugfix to avoid including unnecessary
-Wl,-rpath flags.  It also avoids the warnings on custom_targets with
more than one output.

Reported-by: Laurent Vivier <lvivier@redhat.com>
Reviewed-by: Laurent Vivier <lvivier@redhat.com>
Tested-by: Laurent Vivier <lvivier@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>

usb: usb_packet_map fixes for ehci and xhci.
usb: setup_len fix (CVE-2020-14364).
usb: u2f key support (GSoC).
 * v2: 32bit build fixed.
 * v3: libu2f-emu dependency fixed.

# gpg: Signature made Mon 31 Aug 2020 09:32:49 BST
# gpg:                using RSA key 4CB6D8EED3E87138
# gpg: Good signature from "Gerd Hoffmann (work) <kraxel@redhat.com>" [full]
# gpg:                 aka "Gerd Hoffmann <gerd@kraxel.org>" [full]
# gpg:                 aka "Gerd Hoffmann (private) <kraxel@gmail.com>" [full]
# Primary key fingerprint: A032 8CFF B93A 17A7 9901  FE7D 4CB6 D8EE D3E8 7138

* remotes/kraxel/tags/usb-20200831-pull-request:
  usb: fix setup_len init (CVE-2020-14364)
  usb-host: workaround libusb bug
  hw/usb: Add U2F device autoscan to passthru mode
  hw/usb: Add U2F device check to passthru mode
  scripts: Add u2f-setup-gen script
  docs/qdev-device-use.txt: Add USB U2F key to the QDEV devices examples
  docs/system: Add U2F key to the USB devices examples
  meson: Add U2F key to meson
  hw/usb: Add U2F key emulated mode
  hw/usb: Add U2F key passthru mode
  hw/usb: Add U2F key base class implementation
  hw/usb: Add U2F key base class
  docs: Add USB U2F key device documentation
  hw/usb: Regroup USB HID protocol values
  ehci: drop pointless warn_report for guest bugs.
  hw: ehci: check return value of 'usb_packet_map'
  hw: ehci: destroy sglist in error path
  hw: xhci: check return value of 'usb_packet_map'

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>

Store calculated setup_len in a local variable, verify it, and only
write it to the struct (USBDevice->setup_len) in case it passed the
sanity checks.

This prevents other code (do_token_{in,out} functions specifically)
from working with invalid USBDevice->setup_len values and overrunning
the USBDevice->setup_buf[] buffer.

Fixes: CVE-2020-14364
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Tested-by: Gonglei <arei.gonglei@huawei.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-id: 20200825053636.29648-1-kraxel@redhat.com

libusb_get_device_speed() does not work for
libusb_wrap_sys_device() devices in v1.0.23.

Buglink: https://bugzilla.redhat.com/show_bug.cgi?id=1871090


Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
Message-id: 20200824110057.32089-1-kraxel@redhat.com

This patch adds an autoscan to let u2f-passthru choose the first U2F
device it finds.

The autoscan is performed using libudev with an enumeration of all the
hidraw devices present on the host.

The first device which happens to be a U2F device is taken to do the
passtru.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-13-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

This patchs adds a check to verify that the device passed through the
hidraw property is a U2F device.

The check is done by ensuring that the first values of the report
descriptor (USAGE PAGE and USAGE) correspond to those of a U2F device.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-12-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

This patch adds the script used to generate setup directories, needed
for the device u2f-emulated configuration in directory mode:

    python u2f-setup-gen.py $DIR
    qemu -usb -device u2f-emulated,dir=$DIR

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-11-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-10-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-9-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-8-cesar.belley@lse.epita.fr

[ fixes suggested by paolo ]

Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

This patch adds the U2F key emulated mode.

The emulated mode consists of completely emulating the behavior of a
U2F device through software part. Libu2f-emu is used for that.

The emulated mode is associated with a device inheriting from
u2f-key base.

To work, an emulated U2F device must have differents elements which
can be given in different ways. This is detailed in docs/u2f.txt.

The Ephemeral one is the simplest way to configure, it lets the device
generate all the elements it needs for a single use of the lifetime
of the device:

    qemu -usb -device u2f-emulated

For more information about libu2f-emu see this page:
https://github.com/MattGorko/libu2f-emu

.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-7-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

This patch adds the U2F key pass-through mode.

The pass-through mode consists of passing all requests made from the
guest to the physical security key connected to the host machine and
vice versa.

In addition, the dedicated pass-through allows to have a U2F security key
shared on several guests which is not possible with a simple host device
assignment pass-through.

The pass-through mode is associated with a device inheriting from
u2f-key base.

To work, it needs the path to a U2F hidraw, obtained from the Qemu
command line, and passed by the user:

    qemu -usb -device u2f-passthru,hidraw=/dev/hidrawX

Autoscan and U2F compatibility checking features are given at the end
of the patch series.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-6-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

This patch adds the U2F key base class implementation.

The U2F key base mainly takes care of the HID interfacing with guest.
On the one hand, it retrieves the guest U2FHID packets and transmits
them to the variant associated according to the mode: pass-through
or emulated.
On the other hand, it provides the public API used by its variants to
send U2FHID packets to the guest.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-5-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

This patch adds the specification for the U2F key base class.
Used to group the common characteristics, this device class will be
inherited by its two variants, corresponding to the two modes:
passthrough and emulated

This prepares the U2F devices hierarchy which is as follow:
USB device -> u2f-key -> {u2f-passthru, u2f-emulated}.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-4-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

Add USB U2F key device documentation:
- USB U2F key device
- Building
- Using u2f-emulated
- Using u2f-passthru
- Libu2f-emu

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200826114209.28821-3-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

Group some HID values that are used pretty much everywhere when
dealing with HID devices.

Signed-off-by: César Belley <cesar.belley@lse.epita.fr>
Message-id: 20200812094135.20550-2-cesar.belley@lse.epita.fr
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>