tests/fw_cfg: Free QFWCFG object after qtest has run

We allocate the QFWCFG object previous to run the qtests,
free it once we are finished.

Signed-off-by: Li Qiang <liq3ea@163.com>
Message-Id: <20190424140643.62457-2-liq3ea@163.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
[PMD: Split patch, fill commit description]
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>

tests/libqos: Add pc_fw_cfg_uninit() and use it

The pc_fw_cfg_init() function allocates an IO QFWCFG object.
Add the pc_fw_cfg_uninit() function to deallocate it (and use it).

Signed-off-by: Li Qiang <liq3ea@163.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20190424140643.62457-2-liq3ea@163.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
[PMD: Split patch, fill commit description, call uninit in malloc-pc.c]
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>

tests/libqos: Add io_fw_cfg_uninit() and mm_fw_cfg_uninit()

The mm_fw_cfg_init() allocates a QFWCFG object,
add mm_fw_cfg_uninit() to deallocate it.
Similarly with io_fw_cfg_init(), add io_fw_cfg_uninit().

Signed-off-by: Li Qiang <liq3ea@163.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190424140643.62457-2-liq3ea@163.com>
[PMD: Split patch, filled commit description]
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>

hw/sparc64: Implement fw_cfg_arch_key_name()

Implement fw_cfg_arch_key_name(), which returns the name of a
sparc64-specific key.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-8-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

hw/sparc: Implement fw_cfg_arch_key_name()

Implement fw_cfg_arch_key_name(), which returns the name of a
sparc32-specific key.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-7-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

hw/ppc: Implement fw_cfg_arch_key_name()

Implement fw_cfg_arch_key_name(), which returns the name of a
ppc-specific key.

The fw_cfg device is used by the machine using OpenBIOS:
- 40p
- mac99 (oldworld)
- g3beige (newworld)

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-6-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

hw/i386: Implement fw_cfg_arch_key_name()

Implement fw_cfg_arch_key_name(), which returns the name of a
i386-specific key.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-5-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

hw/i386: Extract fw_cfg definitions to local "fw_cfg.h"

Extract the architecture-specific fw_cfg definitions to "fw_cfg.h".

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-4-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

hw/nvram/fw_cfg: Add fw_cfg_arch_key_name()

Add fw_cfg_arch_key_name() which returns the name of
an architecture-specific key.

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-3-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

hw/nvram/fw_cfg: Add trace events

Add trace events to dump the key content.

Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <20190422195020.1494-2-philmd@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

Merge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-pull-request' into staging

typo fixes, TYPE_XXX usage cleanup, comments update,
virtio-mmio trace functions cleanup

# gpg: Signature made Wed 22 May 2019 17:06:56 BST
# gpg:                using RSA key F30C38BD3F2FBE3C
# gpg: Good signature from "Laurent Vivier <lvivier@redhat.com>" [full]
# gpg:                 aka "Laurent Vivier <laurent@vivier.eu>" [full]
# gpg:                 aka "Laurent Vivier (Red Hat) <lvivier@redhat.com>" [full]
# Primary key fingerprint: CD2F 75DD C8E3 A4DC 2E4F  5173 F30C 38BD 3F2F BE3C

* remotes/vivier2/tags/trivial-branch-pull-request:
  pci: msix: move 'MSIX_CAP_LENGTH' to header file
  vfio: platform: fix a typo
  hw: vfio: drop TYPE_FOO MACRO in VMStateDescription
  vfio: pci: make "vfio-pci-nohotplug" as MACRO
  configure: Fix spelling of sdl-image in --help
  migration: Fix typo in migrate_add_blocker() error message
  roms: List and describe the Makefile 'clean' rule
  roms: Correct the EDK2_BASETOOLS_OPTFLAGS variable description
  hw/virtio/virtio-mmio: Convert DPRINTF to trace and log

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

Merge remote-tracking branch 'remotes/armbru/tags/pull-misc-2019-05-22' into staging

Miscellaneous patches for 2019-05-22

# gpg: Signature made Wed 22 May 2019 14:41:08 BST
# gpg:                using RSA key 3870B400EB918653
# gpg: Good signature from "Markus Armbruster <armbru@redhat.com>" [full]
# gpg:                 aka "Markus Armbruster <armbru@pond.sub.org>" [full]
# Primary key fingerprint: 354B C8B3 D7EB 2A6B 6867  4E5F 3870 B400 EB91 8653

* remotes/armbru/tags/pull-misc-2019-05-22:
  cutils: Simplify how parse_uint() checks for whitespace
  gdbstub: Fix misuse of isxdigit()
  gdbstub: Reject invalid RLE repeat counts
  tests/vhost-user-bridge: Fix misuse of isdigit()
  qemu-bridge-helper: Fix misuse of isspace()

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

Merge remote-tracking branch 'remotes/kraxel/tags/input-20190522-pull-request' into staging

input: add vhost-user-input to contrib.

# gpg: Signature made Wed 22 May 2019 09:26:34 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/input-20190522-pull-request:
  contrib: add vhost-user-input
  libvhost-user: fix -Werror=format= on ppc64
  libvhost-user: fix cast warnings on 32 bits

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

pci: msix: move 'MSIX_CAP_LENGTH' to header file

'MSIX_CAP_LENGTH' is defined in two .c file. Move it
to hw/pci/msix.h file to reduce duplicated code.

CC: qemu-trivial@nongnu.org
Signed-off-by: Li Qiang <liq3ea@163.com>
Message-Id: <20190521151543.92274-5-liq3ea@163.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

vfio: platform: fix a typo

'eventd' should be 'eventfd'.

CC: qemu-trivial@nongnu.org
Signed-off-by: Li Qiang <liq3ea@163.com>
Message-Id: <20190521151543.92274-4-liq3ea@163.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw: vfio: drop TYPE_FOO MACRO in VMStateDescription

It's recommended that VMStateDescription names are decoupled from QOM
type names as the latter may freely change without consideration of
migration compatibility.

Link: https://lists.gnu.org/archive/html/qemu-devel/2018-10/msg02175.html
CC: qemu-trivial@nongnu.org
Signed-off-by: Li Qiang <liq3ea@163.com>
Message-Id: <20190521151543.92274-3-liq3ea@163.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

vfio: pci: make "vfio-pci-nohotplug" as MACRO

The QOMConventions recommends we should use TYPE_FOO
for a TypeInfo's name. Though "vfio-pci-nohotplug" is not
used in other parts, for consistency we should make this change.

CC: qemu-trivial@nongnu.org
Signed-off-by: Li Qiang <liq3ea@163.com>
Message-Id: <20190521151543.92274-2-liq3ea@163.com>
Acked-by: Alex Williamson <alex.williamson@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Eric Auger <eric.auger@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

configure: Fix spelling of sdl-image in --help

Fixes: a442fe2f2b2f20e7be0934277e9400b844b11999
Cc: qemu-trivial@nongnu.org
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20190517183246.11933-1-armbru@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

migration: Fix typo in migrate_add_blocker() error message

Signed-off-by: Greg Kurz <groug@kaod.org>
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <155800428514.543845.17558475870097990036.stgit@bahia.lan>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

roms: List and describe the Makefile 'clean' rule

Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190506141923.12183-3-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

roms: Correct the EDK2_BASETOOLS_OPTFLAGS variable description

In commit 1cab464136b4 we incorrectly described the
EDK2_BASETOOLS_OPTFLAGS can pass CPPFLAGS and CFLAGS
options to the EDK2 build tools, but it only expands
the CFLAGS (not to the CPPFLAGS).
Update the description to be more accurate.

Reported-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190506141923.12183-2-philmd@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

hw/virtio/virtio-mmio: Convert DPRINTF to trace and log

Use traces for debug message and qemu_log_mask for errors.

Signed-off-by: Boxuan Li <liboxuan@connect.hku.hk>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Message-Id: <20190503154424.73933-1-liboxuan@connect.hku.hk>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

cutils: Simplify how parse_uint() checks for whitespace

Use qemu_isspace() so we don't have to cast to unsigned char.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190514180311.16028-7-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>

gdbstub: Fix misuse of isxdigit()

gdb_read_byte() passes its @ch argument to isxdigit().  Undefined
behavior when the value is negative.  Two callers:

* gdb_chr_receive() passes an uint8_t value.  Safe.

* gdb_handlesig() a char value.  Unsafe.  Not a security issue,
  because the characters come from the gdb client, which is trusted.

The obvious fix would be casting @ch to unsigned char.  But note that
gdb_read_byte() already casts @ch to uint8_t in many places.  Uses of
@ch without such a cast:

(1) Compare to a character constant with == or !=

(2) s->linesum += ch

(3) Store ch or ch ^ 0x20 into s->line_buf[]

(4) Check for invalid RLE count:
    ch < ' ' || ch == '#' || ch == '$' || ch > 126

(5) Pass to isxdigit()

(6) Pass to fromhex()

Change the parameter type from int to uint8_t, and drop the now
redundant casts.  Affects the above uses as follows:

(1) No change: the character constants are all non-negative.

(2) Effectively no change: we only ever use s->linesum & 0xff, and
    s->linesum is int.

(3) No change: s->line_buf[] is char[].

(4) No change.

(5) Avoid undefined behavior.

(6) No change: only reached when isxdigit(ch)

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190514180311.16028-5-armbru@redhat.com>

gdbstub: Reject invalid RLE repeat counts

"Debugging with GDB / Appendix E GDB Remote Serial Protocol /
Overview" specifies "The printable characters '#' and '$' or with a
numeric value greater than 126 must not be used."  gdb_read_byte()
only rejects values < 32.  This is wrong.  Impact depends on the caller:

* gdb_handlesig() passes a char.  Incorrectly accepts '#', '$' and
  '\127'.

* gdb_chr_receive() passes an uint8_t.  Additionally accepts
  characters with the most-significant bit set.

Correct the validity check to match the specification.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190514180311.16028-4-armbru@redhat.com>

tests/vhost-user-bridge: Fix misuse of isdigit()

vubr_set_host() passes char values to isdigit().  Undefined behavior
when the value is negative.

Fix by using qemu_isdigit() instead.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190514180311.16028-3-armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
[Missing #include "qemu-common.h" fixed]

qemu-bridge-helper: Fix misuse of isspace()

parse_acl_file() passes char values to isspace().  Undefined behavior
when the value is negative.  Not a security issue, because the
characters come from trusted $prefix/etc/qemu/bridge.conf and the
files it includes.

Furthermore, isspace()'s locale-dependence means qemu-bridge-helper
uses the user's locale for parsing $prefix/etc/bridge.conf.  Feels
wrong.

Use g_ascii_isspace() instead.  This fixes the undefined behavior, and
makes parsing of $prefix/etc/bridge.conf locale-independent.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190514180311.16028-2-armbru@redhat.com>

contrib: add vhost-user-input

Add a vhost-user input backend example, based on virtio-input-host
device. It takes an evdev path as argument, and can be associated with
a vhost-user-input device via a UNIX socket:

$ vhost-user-input -p /dev/input/eventX -s /tmp/vui.sock

$ qemu ... -chardev socket,id=vuic,path=/tmp/vui.sock
  -device vhost-user-input-pci,chardev=vuic

This example is intentionally not included in $TOOLS, and not
installed by default.

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-id: 20190514104126.6294-4-marcandre.lureau@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

libvhost-user: fix -Werror=format= on ppc64

That should fix the following warning:

/home/pm215/qemu/contrib/libvhost-user/libvhost-user.c: In function
‘vu_set_mem_table_exec_postcopy’:
/home/pm215/qemu/contrib/libvhost-user/libvhost-user.c:666:9: error:
format ‘%llx’ expects argument of type ‘long long unsigned int’, but
argument 5 has type ‘__u64’ [-Werror=format=]
         DPRINT("%s: region %d: Registered userfault for %llx + %llx\n",
         ^
/home/pm215/qemu/contrib/libvhost-user/libvhost-user.c:666:9: error:
format ‘%llx’ expects argument of type ‘long long unsigned int’, but
argument 6 has type ‘__u64’ [-Werror=format=]
cc1: all warnings being treated as errors

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190514104126.6294-3-marcandre.lureau@redhat.com

{ kraxel: s/PRIu64/PRIx64/ ]

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

libvhost-user: fix cast warnings on 32 bits

Fixes warnings:
 warning: cast to pointer from integer of different size
 [-Wint-to-pointer-cast]

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190514104126.6294-2-marcandre.lureau@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>

Merge remote-tracking branch 'remotes/ehabkost/tags/x86-next-pull-request' into staging

x86 MDS feature flags

md-clear and mds-no feature flags, for detection and mitigation
of MDS vulnerabilities (CVE-2018-12126, CVE-2018-12127,
CVE-2018-12130, CVE-2019-11091).

# gpg: Signature made Tue 21 May 2019 19:42:43 BST
# gpg:                using RSA key 2807936F984DC5A6
# gpg: Good signature from "Eduardo Habkost <ehabkost@redhat.com>" [full]
# Primary key fingerprint: 5A32 2FD5 ABC4 D3DB ACCF  D1AA 2807 936F 984D C5A6

* remotes/ehabkost/tags/x86-next-pull-request:
  target/i386: add MDS-NO feature
  docs: recommend use of md-clear feature on all Intel CPUs
  target/i386: define md-clear bit

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

target/i386: add MDS-NO feature

Microarchitectural Data Sampling is a hardware vulnerability which allows
unprivileged speculative access to data which is available in various CPU
internal buffers.

Some Intel processors use the ARCH_CAP_MDS_NO bit in the
IA32_ARCH_CAPABILITIES
MSR to report that they are not vulnerable, make it available to guests.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20190516185320.28340-1-pbonzini@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

docs: recommend use of md-clear feature on all Intel CPUs

Update x86 CPU model guidance to recommend that the md-clear feature is
manually enabled with all Intel CPU models, when supported by the host
microcode.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190515141011.5315-3-berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

target/i386: define md-clear bit

md-clear is a new CPUID bit which is set when microcode provides the
mechanism to invoke a flush of various exploitable CPU buffers by invoking
the VERW instruction.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20190515141011.5315-2-berrange@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>

Merge remote-tracking branch 'remotes/cohuck/tags/s390x-20190521-3' into staging

s390x update:
- have the bios tolerate bootmap signature entries
- next chunk of vector instruction support in tcg
- a headers update against Linux 5.2-rc1
- add more facilities and gen15 machines to the cpu model

# gpg: Signature made Tue 21 May 2019 16:09:35 BST
# gpg:                using RSA key C3D0D66DC3624FF6A8C018CEDECF6B93C6F02FAF
# gpg:                issuer "cohuck@redhat.com"
# gpg: Good signature from "Cornelia Huck <conny@cornelia-huck.de>" [unknown]
# gpg:                 aka "Cornelia Huck <huckc@linux.vnet.ibm.com>" [full]
# gpg:                 aka "Cornelia Huck <cornelia.huck@de.ibm.com>" [full]
# gpg:                 aka "Cornelia Huck <cohuck@kernel.org>" [unknown]
# gpg:                 aka "Cornelia Huck <cohuck@redhat.com>" [unknown]
# Primary key fingerprint: C3D0 D66D C362 4FF6 A8C0  18CE DECF 6B93 C6F0 2FAF

* remotes/cohuck/tags/s390x-20190521-3: (55 commits)
  s390x/cpumodel: wire up 8561 and 8562 as gen15 machines
  s390x/cpumodel: add gen15 defintions
  s390x/cpumodel: add Deflate-conversion facility
  s390x/cpumodel: enhanced sort facility
  s390x/cpumodel: vector enhancements
  s390x/cpumodel: msa9 facility
  s390x/cpumodel: Miscellaneous-Instruction-Extensions Facility 3
  s390x/cpumodel: ignore csske for expansion
  linux headers: update against Linux 5.2-rc1
  update-linux-headers: handle new header file
  s390x/tcg: Implement VECTOR TEST UNDER MASK
  s390x/tcg: Implement VECTOR SUM ACROSS WORD
  s390x/tcg: Implement VECTOR SUM ACROSS QUADWORD
  s390x/tcg: Implement VECTOR SUM ACROSS DOUBLEWORD
  s390x/tcg: Implement VECTOR SUBTRACT WITH BORROW COMPUTE BORROW INDICATION
  s390x/tcg: Implement VECTOR SUBTRACT WITH BORROW INDICATION
  s390x/tcg: Implement VECTOR SUBTRACT COMPUTE BORROW INDICATION
  s390x/tcg: Implement VECTOR SUBTRACT
  s390x/tcg: Implement VECTOR SHIFT RIGHT LOGICAL *
  s390x/tcg: Implement VECTOR SHIFT RIGHT ARITHMETIC
  ...

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

s390x/cpumodel: wire up 8561 and 8562 as gen15 machines

8561 and 8562 will be gen15 machines. There is no name yet, let us use
gen15a and gen15b as base name. Later on we can provide aliases with
the proper name.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190429090250.7648-10-borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: add gen15 defintions

add several new features (msa9, sort, deflate, additional vector
instructions, new general purpose instructions) to generation 15.

Also disable csske and bpb from the default and base models >=15.
This will allow to migrate gen15 machines to future machines that
do not have these features.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190429090250.7648-9-borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: add Deflate-conversion facility

add the deflate conversion facility.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190429090250.7648-8-borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: enhanced sort facility

add the enhanced sort facility.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190429090250.7648-7-borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: vector enhancements

Add vector enhancements to the cpu model.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190429090250.7648-6-borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: msa9 facility

Provide the MSA9 facility (stfle.155). This also contains pckmo
subfunctions for key wrapping. Keep them in a separate group to disable
those as a block if necessary. This is for example needed when disabling
key wrapping via the HMC.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190429090250.7648-5-borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: Miscellaneous-Instruction-Extensions Facility 3

Provide the "Miscellaneous-Instruction-Extensions Facility 3" via
stfle.61.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190429090250.7648-4-borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/cpumodel: ignore csske for expansion

csske will be removed in a future machine. Ignore it for expanding the
cpu model. Otherwise qemu falls back to z9.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Cc: qemu-stable@nongnu.org
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190429090250.7648-3-borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

linux headers: update against Linux 5.2-rc1

commit a188339ca5a396acc588e5851ed7e19f66b0ebd9

Signed-off-by: Cornelia Huck <cohuck@redhat.com>

update-linux-headers: handle new header file

We need to copy sve_context.h for aarch64.

Signed-off-by: Cornelia Huck <cohuck@redhat.com>

Merge remote-tracking branch 'remotes/mst/tags/for_upstream' into staging

pci, pc, virtio: features, fixes

reconnect for vhost blk
tests for UEFI
misc other stuff

Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
# gpg: Signature made Tue 21 May 2019 14:41:32 BST
# gpg:                using RSA key 281F0DB8D28D5469
# gpg: Good signature from "Michael S. Tsirkin <mst@kernel.org>" [full]
# gpg:                 aka "Michael S. Tsirkin <mst@redhat.com>" [full]
# Primary key fingerprint: 0270 606B 6F3C DF3D 0B17  0970 C350 3912 AFBE 8E67
#      Subkey fingerprint: 5D09 FD08 71C8 F85B 94CA  8A0D 281F 0DB8 D28D 5469

* remotes/mst/tags/for_upstream: (34 commits)
  tests: acpi: print error unable to dump ACPI table during rebuild
  tests: acpi: refactor rebuild-expected-aml.sh to dump ACPI tables for a specified list of targets
  tests: acpi: allow to override default accelerator
  tests: acpi: ignore SMBIOS tests when UEFI firmware is used
  tests: acpi: add a way to start tests with UEFI firmware
  tests: acpi: add acpi_find_rsdp_address_uefi() helper
  tests: acpi: move boot_sector_init() into x86 tests branch
  tests: acpi: skip FACS table if board uses hw reduced ACPI profile
  tests: acpi: fetch X_DSDT if pointer to DSDT is 0
  tests: acpi: make pointer to RSDP 64bit
  tests: acpi: make RSDT test routine handle XSDT
  tests: acpi: make acpi_fetch_table() take size of fetched table pointer
  tests: acpi: rename acpi_parse_rsdp_table() into acpi_fetch_rsdp_table()
  pci: Simplify pci_bus_is_root()
  pcie: Remove redundant test in pcie_mmcfg_data_{read,write}()
  libvhost-user: fix bad vu_log_write
  hw/arm/virt-acpi-build: pass AcpiMcfgInfo to build_mcfg()
  i386, acpi: remove mcfg_ prefix in AcpiMcfgInfo members
  hw/arm/virt-acpi-build: remove unnecessary variable mcfg_start
  do not call vhost_net_cleanup() on running net from char user event
  ...

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

tests: acpi: print error unable to dump ACPI table during rebuild

Instead of just asserting print the error that lead to assert first.
While at it move assert into rebuild branch, which removes redundant
check done in case of !rebuild branch is taken (the later is taken
care of by g_assert_no_error).

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <1556808723-226478-16-git-send-email-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: refactor rebuild-expected-aml.sh to dump ACPI tables for a specified list of targets

Make initial list contain x86_64.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <1556808723-226478-15-git-send-email-imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2019-05-21' into staging

- qtest patches to get rid of the global_qtest variable in more tests
- some iotests patches that have multiple reviews and thus are ready to go

# gpg: Signature made Tue 21 May 2019 11:40:31 BST
# gpg:                using RSA key 2ED9D774FE702DB5
# gpg: Good signature from "Thomas Huth <th.huth@gmx.de>" [full]
# gpg:                 aka "Thomas Huth <thuth@redhat.com>" [full]
# gpg:                 aka "Thomas Huth <huth@tuxfamily.org>" [full]
# gpg:                 aka "Thomas Huth <th.huth@posteo.de>" [unknown]
# Primary key fingerprint: 27B8 8847 EEE0 2501 18F3  EAB9 2ED9 D774 FE70 2DB5

* remotes/huth-gitlab/tags/pull-request-2019-05-21:
  tests/qemu-iotests: Remove the "_supported_os Linux" line from many tests
  cirrus / travis: Add gnu-sed and bash for macOS and FreeBSD
  tests/qemu-iotests: Do not hard-code the path to bash
  tests/qemu-iotests/check: Pick a default machine if necessary
  tests/qemu-iotests/005: Add a sanity check for large sparse file support
  tests/hd-geo-test: Use qtest_init() instead of qtest_start()
  tests/device-introspect: Use qtest_init() instead of qtest_start()
  tests/qom-test: Use qtest_init() instead of qtest_start()
  tests/numa-test: Use qtest_init() instead of qtest_start()
  tests/q35-test: Make test independent of global_qtest
  tests/libqos: Get rid of global_qtest dependency in qvring_init()
  tests/libqtest: Fix description of qtest_vinitf() and qtest_initf()
  tests/libqtest: Remove unused global_qtest-related wrapper functions

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

Merge remote-tracking branch 'remotes/xtensa/tags/20190520-xtensa' into staging

target/xtensa: SR reorganization and options for modern cores

Reorganize special register handling to support configurations with
conflicting SR definitions.

Implement options used by the modern xtensa cores:
- memory protection unit;
- block prefetch;
- exclusive access

Add special register definitions and IRQ types for ECC/parity,
gather/scatter and IDMA.

# gpg: Signature made Mon 20 May 2019 18:53:05 BST
# gpg:                using RSA key 2B67854B98E5327DCDEB17D851F9CC91F83FA044
# gpg:                issuer "jcmvbkbc@gmail.com"
# gpg: Good signature from "Max Filippov <filippov@cadence.com>" [unknown]
# gpg:                 aka "Max Filippov <max.filippov@cogentembedded.com>" [full]
# gpg:                 aka "Max Filippov <jcmvbkbc@gmail.com>" [full]
# Primary key fingerprint: 2B67 854B 98E5 327D CDEB  17D8 51F9 CC91 F83F A044

* remotes/xtensa/tags/20190520-xtensa:
  target/xtensa: implement exclusive access option
  target/xtensa: update list of exception causes
  target/xtensa: implement block prefetch option opcodes
  target/xtensa: implement DIWBUI.P opcode
  target/xtensa: implement MPU option
  target/xtensa: add parity/ECC option SRs
  target/xtensa: define IDMA and gather/scatter IRQ types
  target/xtensa: make internal MMU functions static
  target/xtensa: get rid of centralized SR properties

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

tests/qemu-iotests: Remove the "_supported_os Linux" line from many tests

A lot of tests run fine on FreeBSD and macOS, too - the limitation
to Linux here was likely just copied-and-pasted from other tests.
Thus remove the "_supported_os Linux" line from tests that run
successful in our CI pipelines on FreeBSD and macOS.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190502084506.8009-6-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

cirrus / travis: Add gnu-sed and bash for macOS and FreeBSD

We are going to enable the qemu-iotests during "make check" again,
and for running the iotests, we need bash and gnu-sed.

Reviewed-by: Li-Wen Hsu <lwhsu@freebsd.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190502084506.8009-5-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/qemu-iotests: Do not hard-code the path to bash

bash is installed in a different directory on non-Linux systems like
FreeBSD. Do not hard-code /bin/bash here so that the tests can run
there, too.

Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190502084506.8009-4-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/qemu-iotests/check: Pick a default machine if necessary

qemu-system-arm, qemu-system-aarch64 and qemu-system-tricore do not have
a default machine, so when running the qemu-iotests with such a binary,
lots of tests are failing. Fix it by picking a default machine in the
"check" script instead.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Christian Borntraeger <borntraeger@de.ibm.com>
Message-Id: <20190502084506.8009-3-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/qemu-iotests/005: Add a sanity check for large sparse file support

"check -raw 005" fails when running on certain filesystems - these do not
support such large sparse files. Use the same check as in test 220 to
skip the test in this case.

Suggested-by: Eric Blake <eblake@redhat.com>
Message-Id: <20190502084506.8009-2-thuth@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/hd-geo-test: Use qtest_init() instead of qtest_start()

qtest_start() + qtest_end() should be avoided, since they use the
global_qtest variable that we want to get rid of in the long run.
Use qtest_init() and qtest_quit() instead.

Message-Id: <20190515174328.16361-7-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/device-introspect: Use qtest_init() instead of qtest_start()

qtest_start() + qtest_end() should be avoided, since they use the
global_qtest variable that we want to get rid of in the long run.
Use qtest_init() and qtest_quit() instead.

Message-Id: <20190515174328.16361-6-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/qom-test: Use qtest_init() instead of qtest_start()

qtest_start() + qtest_end() should be avoided, since they use the
global_qtest variable that we want to get rid of in the long run.
Use qtest_init() and qtest_quit() instead.

Message-Id: <20190515174328.16361-5-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/numa-test: Use qtest_init() instead of qtest_start()

qtest_start() + qtest_end() should be avoided, since they use the
global_qtest variable that we want to get rid of in the long run.
Use qtest_init() and qtest_quit() instead.

Message-Id: <20190515174328.16361-4-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/q35-test: Make test independent of global_qtest

Use a local QTestState variable, so that we can finally get rid
of the undesired global_qtest variable in this file, too.

Message-Id: <20190515174328.16361-3-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/libqos: Get rid of global_qtest dependency in qvring_init()

Library functions should not depend on global_qtest functions like
writew() and writeq(), so that they can also be used in tests that
deal with multiple QTestStates at the same time (like migration tests).

Message-Id: <20190515174328.16361-2-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/libqtest: Fix description of qtest_vinitf() and qtest_initf()

These functions are convenience wrappers of qtest_init() and not of
qtest_start().

Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190513154759.24973-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/libqtest: Remove unused global_qtest-related wrapper functions

A bunch of the wrapper functions that use global_qtest are not used
anymore. Remove them to avoid that they are used in new code again.

Reviewed-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Message-Id: <20190510052239.21947-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests: acpi: allow to override default accelerator

By default test cases were run with 'kvm:tcg' accelerators to speed up
tests execution. While it works for x86, were change of accelerator
doesn't affect ACPI tables, the approach doesn't works for ARM usecase
though.

In arm/virt case, KVM mode requires using 'host' cpu model, which
isn't available in TCG mode. That could be worked around with 'max'
cpu model, which works both for KVM and TCG. However in KVM mode it
is necessary to specify matching GIC version, which also could use
'max' value to automatically pick GIC version suitable for host's CPU.
Depending on host cpu type, different GIC versions would be used,
which in turn leads to different ACPI tables (APIC) generated.
As result while comparing with reference blobs, test would fail if
host's GIC version won't match the version on the host where
reference blobs where generated.

Let's keep testing simple for now and allow ARM tests run in TCG only
mode. To do so introduce 'accel' parameter in test configuration, so
test case could override default "kvm:tcg" with accelerator of choice.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1556808723-226478-12-git-send-email-imammedo@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: ignore SMBIOS tests when UEFI firmware is used

once FW provides a pointer to SMBIOS entry point like it does for
RSDP it should be possible to enable this one the same way.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <1556808723-226478-11-git-send-email-imammedo@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: add a way to start tests with UEFI firmware

For testcase to use UEFI firmware, one needs to provide and specify
firmware and varstore blob names in test_data { uefi_fl1, uefi_fl2 }
fields respectively and RAM start address plus size where to look for
test structure signature. Additionally testcase should specify
bootable cdrom image from uefi-boot-images with EFI test utility.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Message-Id: <1556808723-226478-10-git-send-email-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: add acpi_find_rsdp_address_uefi() helper

introduce UEFI specific counterpart to acpi_find_rsdp_address()
that will help to find RSDP address when [OA]VMF is used as
firmware. It requires guest firmware or other guest app to place
1Mb aligned UefiTestSupport structure (defined in this patch)
in RAM with UefiTestSupport::signature_guid set to
AB87A6B1-2034-BDA0-71BD-375007757785
For test app details see commit
  (09a274d82f tests: introduce "uefi-test-tools" with the BiosTablesTest UEFI app)

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Message-Id: <1556808723-226478-9-git-send-email-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: move boot_sector_init() into x86 tests branch

boot_sector_init() won't be used by arm/virt board, so move it from
global scope to x86 branch that uses it.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <1556808723-226478-8-git-send-email-imammedo@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: skip FACS table if board uses hw reduced ACPI profile

If FADT has HW_REDUCED_ACPI flag set, do not attempt to fetch
FACS as it's not provided by the board.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Message-Id: <1556808723-226478-7-git-send-email-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: fetch X_DSDT if pointer to DSDT is 0

that way it would be possible to test a DSDT pointed by
64bit X_DSDT field in FADT.

PS:
it will allow to enable testing arm/virt board, which sets
only newer X_DSDT field.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1556808723-226478-6-git-send-email-imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: make pointer to RSDP 64bit

In case of UEFI, RSDP doesn't have to be located in lowmem,
it could be placed at any address. Make sure that test won't
break if it is placed above the first 4Gb of address space.

PS:
While at it cleanup some local variables as we don't really
need them.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <1556808723-226478-5-git-send-email-imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: make RSDT test routine handle XSDT

If RSDP revision is more than 0 fetch table pointed by XSDT
and fallback to legacy RSDT table otherwise.

While at it drop unused acpi_get_xsdt_address().

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1556808723-226478-4-git-send-email-imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: make acpi_fetch_table() take size of fetched table pointer

Currently acpi_fetch_table() assumes 32 bit size of table pointer
in ACPI tables. However X_foo variants are 64 bit, prepare
acpi_fetch_table() to handle both by adding an argument
for addr_ptr pointed entry size. Follow up commits will use that
to read XSDT and X_foo entries in ACPI tables.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1556808723-226478-3-git-send-email-imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

tests: acpi: rename acpi_parse_rsdp_table() into acpi_fetch_rsdp_table()

so name would reflect what the function does

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Message-Id: <1556808723-226478-2-git-send-email-imammedo@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

pci: Simplify pci_bus_is_root()

pci_bus_is_root() currently relies on a method in the PCIBusClass.
But it's always known if a PCI bus is a root bus when we create it, so
using a dynamic method is overkill.

This replaces it with an IS_ROOT bit in a new flags field, which is set on
root buses and otherwise clear.  As a bonus this removes the special
is_root logic from pci_expander_bridge, since it already creates its bus
as a root bus.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Reviewed-by: Marcel Apfelbaum <marcel@redhat.com>
Reviewed-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Greg Kurz <groug@kaod.org>
Message-Id: <20190424041959.4087-3-david@gibson.dropbear.id.au>

pcie: Remove redundant test in pcie_mmcfg_data_{read,write}()

These functions have an explicit test for accesses above the device's
config size.  But pci_host_config_{read,write}_common() which they're
about to call already have checks against the config space limit and
do the right thing.  So, remove the redundant tests.

Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
Message-Id: <20190424041959.4087-2-david@gibson.dropbear.id.au>

libvhost-user: fix bad vu_log_write

Mark dirty as page, the step of each call is 1.

Signed-off-by: Li Feng <fengli@smartx.com>
Message-Id: <20190420091016.213160-1-fengli@smartx.com>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

hw/arm/virt-acpi-build: pass AcpiMcfgInfo to build_mcfg()

To build MCFG, two information is necessary:

    * bus number
    * base address

Abstract these two information to AcpiMcfgInfo so that build_mcfg and
build_mcfg_q35 will have the same declaration.

Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <20190419003053.8260-5-richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

i386, acpi: remove mcfg_ prefix in AcpiMcfgInfo members

This is obvious the member in AcpiMcfgInfo describe MCFG's property.

Remove the mcfg_ prefix.

Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
Suggested-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190419003053.8260-4-richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

hw/arm/virt-acpi-build: remove unnecessary variable mcfg_start

mcfg_start points to the start of MCFG table and is used in
build_header. While this information could be derived from mcfg.

This patch removes the unnecessary variable mcfg_start.

Signed-off-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190419003053.8260-3-richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

do not call vhost_net_cleanup() on running net from char user event

Buglink: https://launchpad.net/bugs/1823458
Currently, a user CHR_EVENT_CLOSED event will cause net_vhost_user_event()
to call vhost_user_cleanup(), which calls vhost_net_cleanup() for all
its queues.  However, vhost_net_cleanup() must never be called like
this for fully-initialized nets; when other code later calls
vhost_net_stop() - such as from virtio_net_vhost_status() - it will try
to access the already-cleaned-up fields and fail with assertion errors
or segfaults.

The vhost_net_cleanup() will eventually be called from
qemu_cleanup_net_client().

Signed-off-by: Dan Streetman <ddstreet@canonical.com>
Message-Id: <20190416184624.15397-3-dan.streetman@canonical.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

q35: acpi: do not create dummy MCFG table

Dummy table (with signature "QEMU") creation came from original SeaBIOS
codebase. And QEMU would have to keep it around if there were Q35 machine
that depended on keeping ACPI tables blob constant size. Luckily there
were no versioned Q35 machine types before commit:
  (since 2.3) a1666142db acpi-build: make ROMs RAM blocks resizeable
which obsoleted need to keep ACPI tables blob the same size on source/destination.

Considering the 1st versioned machine is pc-q35-2.4, the dummy table
is not really necessary and it's safe to drop it without breaking
cross version migration in both directions unconditionally.

Signed-off-by: Igor Mammedov <imammedo@redhat.com>
Message-Id: <1554822037-329838-1-git-send-email-imammedo@redhat.com>
Reviewed-by: Wei Yang <richardw.yang@linux.intel.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

acpi/pcihp: Add a few more trace points related to unplug

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190402161900.7374-4-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>

acpi/pcihp: Convert debug printf()s to trace events

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190402161900.7374-3-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>

acpi/piix4: Convert debug printf()s to trace events

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190402161900.7374-2-armbru@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Igor Mammedov <imammedo@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>

contrib/vhost-user-blk: enable inflight I/O tracking

This patch enables inflight I/O tracking for
vhost-user-blk backend so that we could restart it safely.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-8-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

vhost-user-blk: Add support to reconnect backend

Since we now support the message VHOST_USER_GET_INFLIGHT_FD
and VHOST_USER_SET_INFLIGHT_FD. The backend is able to restart
safely because it can track inflight I/O in shared memory.
This patch allows qemu to reconnect the backend after
connection closed.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Ni Xun <nixun@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-7-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

vhost-user-blk: Add return value for vhost_user_blk_start()

Add a return value for vhost_user_blk_start() to check whether
we start vhost-user backend successfully or not.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-6-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

vhost-user-blk: Only start vhost-user backend with the first kick

We should only start vhost-user backend at the first kick for
virtio 1.0 transitional devices.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-5-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

vhost-user-blk: Use started flag in vhost_user_blk_set_status()

Use started flag in vhost_user_blk_set_status() to decide if
starting vhost-user backend or not.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-4-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

virtio: Use started flag in virtio_vmstate_change()

Currently, we use DRIVER_OK status bit to check whether guest
driver has started the device in virtio_vmstate_change(). But it's
not the case for virtio 1.0 transitional devices. If migration completes
between kicking virtqueue and setting VIRTIO_CONFIG_S_DRIVER_OK, guest
may be hung. So here we use started flag to check guest state instead.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-3-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

virtio: Introduce started flag to VirtioDevice

The virtio 1.0 transitional devices support driver uses the device
before setting the DRIVER_OK status bit. So we introduce a started
flag to indicate whether driver has started the device or not.

Signed-off-by: Xie Yongji <xieyongji@baidu.com>
Signed-off-by: Zhang Yu <zhangyu31@baidu.com>
Message-Id: <20190320112646.3712-2-xieyongji@baidu.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

docs: reST-ify vhost-user documentation

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20190315180735.13096-1-marcandre.lureau@redhat.com>
Reviewed-by: Jens Freimann <jfreimann@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

Revert "globals: Allow global properties to be optional"

This reverts commit d7741743f4f3d2683d1bb6938f88dc0167c21afa.

Relying on setting properties on parents types which may not
be relevant to certain sub-classes had unexpected side-effects
causing bugs in device config defaults. It is preferrable to
be explicit about which devices get which properties, even if
this needs repetition.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190215103239.28640-3-berrange@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

hw: report invalid disable-legacy|modern usage for virtio-1-only devs

A number of virtio devices (gpu, crypto, mouse, keyboard, tablet) only
support the virtio-1 (aka modern) mode. Currently if the user launches
QEMU, setting those devices to enable legacy mode, QEMU will silently
create them in modern mode, ignoring the user's (mistaken) request.

This patch introduces proper data validation so that an attempt to
configure a virtio-1-only devices in legacy mode gets reported as an
error to the user.

Checking this required introduction of a new field to explicitly track
what operating model is to be used for a device, separately from the
disable_modern and disable_legacy fields that record the user's
requested configuration.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190215103239.28640-2-berrange@redhat.com>
Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>

Merge remote-tracking branch 'remotes/kevin/tags/for-upstream' into staging

Block layer patches:

- block: AioContext management, part 1
- qmp: forbid qmp_cont in RUN_STATE_FINISH_MIGRATE
- nvme: fix copy direction in DMA reads going to CMB
- file-posix: Fix block status for unaligned raw images with O_DIRECT
- file-posix: Fix xfs_write_zeroes() after EOF
- Documentation and iotests improvements

# gpg: Signature made Mon 20 May 2019 16:12:38 BST
# gpg:                using RSA key 7F09B272C88F2FD6
# gpg: Good signature from "Kevin Wolf <kwolf@redhat.com>" [full]
# Primary key fingerprint: DC3D EB15 9A9A F95D 3D74  56FE 7F09 B272 C88F 2FD6

* remotes/kevin/tags/for-upstream: (24 commits)
  iotests: Make 245 faster and more reliable
  iotests.py: Fix VM.run_job
  iotests.py: Let assert_qmp() accept an array
  block: Improve "Block node is read-only" message
  qemu-img.texi: Describe human-readable info output
  qemu-img.texi: Be specific about JSON object types
  iotests: Test unaligned raw images with O_DIRECT
  block/file-posix: Unaligned O_DIRECT block-status
  test-block-iothread: Test AioContext propagation for block jobs
  blockjob: Remove AioContext notifiers
  blockjob: Propagate AioContext change to all job nodes
  block: Add blk_set_allow_aio_context_change()
  block: Implement .(can_)set_aio_ctx for BlockBackend
  test-block-iothread: Test AioContext propagation through the tree
  block: Propagate AioContext change to parents
  block: Move recursion to bdrv_set_aio_context()
  block: Make bdrv_attach/detach_aio_context() static
  block: Add bdrv_try_set_aio_context()
  nvme: fix copy direction in DMA reads going to CMB
  iotest: fix 169: do not run qmp_cont in RUN_STATE_FINISH_MIGRATE
  ...

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

iotests: Make 245 faster and more reliable

Sometimes, 245 fails for me because some stream job has already finished
while the test expects it to still be active.  (With -c none, it fails
basically every time.) The most reliable way to fix this is to simply
set auto_finalize=false so the job will remain in the block graph as
long as we need it.  This allows us to drop the rate limiting, too,
which makes the test faster.

The only problem with this is that there is a single place that yields a
different error message depending on whether the stream job is still
copying data (so COR is enabled) or not (COR has been disabled, but the
job still has the WRITE_UNCHANGED permission on the target node).  We
can easily address that by expecting either error message.

Note that we do not need auto_finalize=false (or rate limiting) for the
active commit job, because It never completes without an explicit
block-job-complete anyway.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

iotests.py: Fix VM.run_job

log() is in the current module, there is no need to prefix it.  In fact,
doing so may make VM.run_job() unusable in tests that never use
iotests.log() themselves.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

iotests.py: Let assert_qmp() accept an array

Sometimes we cannot tell which error message qemu will emit, and we do
not care.  With this change, we can then just pass an array of all
possible messages to assert_qmp() and it will choose the right one.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>

block: Improve "Block node is read-only" message

This message does not make any sense when it appears as the response to
making an R/W node read-only.  We should detect that case and emit a
different message, then.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Reviewed-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>