target/s390x: Restrict system-mode declarations

As these declarations are restricted to !CONFIG_USER_ONLY in
helper.c, only declare them when system-mode emulation is used.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200526172427.17460-8-f4bug@amsat.org>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

target/s390x/helper: Clean ifdef'ry

All this code is guarded checking CONFIG_USER_ONLY definition.
Drop the duplicated checks.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200526172427.17460-7-f4bug@amsat.org>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

target/s390x: Only compile decode_basedisp() on system-mode

The decode_basedisp*() methods are only used in ioinst.c,
which is only build in system-mode emulation.

I/O instructions are privileged, and other S instructions
are decoded elsewhere.

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Cornelia Huck <cohuck@redhat.com>
Message-Id: <20200526172427.17460-6-f4bug@amsat.org>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

MAINTAINERS: add Thomas as additional s390x maintainer

...because two people are better than one.

Signed-off-by: Cornelia Huck <cohuck@redhat.com>
Acked-by: Thomas Huth <thuth@redhat.com>
Cc: Thomas Huth <thuth@redhat.com>
Message-Id: <20200525155855.225564-1-cohuck@redhat.com>

docs/s390x: document vfio-ccw

Add a basic example for passing a dasd via vfio-ccw.

Message-Id: <20200518075522.97643-1-cohuck@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

vfio-ccw: allow non-prefetch ORBs

Remove the explicit prefetch check when using vfio-ccw devices.
This check does not trigger in practice as all Linux channel programs
are intended to use prefetch.

Newer Linux kernel versions do not require to force the PFCH flag with
vfio-ccw devices anymore.

Signed-off-by: Jared Rossi <jrossi@linux.ibm.com>
Reviewed-by: Eric Farman <farman@linux.ibm.com>
Message-Id: <20200512181535.18630-2-jrossi@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

docs/s390x: document 3270

Add some basic info how to use 3270 devices.

Message-Id: <20200515151518.83950-3-cohuck@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

docs/s390x: document the virtual css

Add some hints about "devno" rules.

Message-Id: <20200515151518.83950-2-cohuck@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x: pv: Fix KVM_PV_PREP_RESET command wrapper name

s390_pv_perf_clear_reset() is not a very helpful name since that
function needs to be called for a normal and a clear reset via
diag308.

Let's instead name it s390_pv_prep_reset() which reflects the purpose
of the function a bit better.

Signed-off-by: Janosch Frank <frankja@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200505124159.24099-1-frankja@linux.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

s390x/kvm: help valgrind in several places

We need some little help in the code to reduce the valgrind noise.
This patch does this with some designated initializers for the cpu
model features and subfunctions.

Signed-off-by: Christian Borntraeger <borntraeger@de.ibm.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20200429074201.100924-1-borntraeger@de.ibm.com>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>

Merge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging

Pull request

# gpg: Signature made Fri 05 Jun 2020 10:47:27 BST
# gpg:                using RSA key 8695A8BFD3F97CDAAC35775A9CA4ABB381AB73C8
# gpg: Good signature from "Stefan Hajnoczi <stefanha@redhat.com>" [full]
# gpg:                 aka "Stefan Hajnoczi <stefanha@gmail.com>" [full]
# Primary key fingerprint: 8695 A8BF D3F9 7CDA AC35  775A 9CA4 ABB3 81AB 73C8

* remotes/stefanha/tags/block-pull-request:
  block: Factor out bdrv_run_co()
  exec: Rename qemu_ram_writeback() as qemu_ram_msync()
  hw/block: Let the NVMe emulated device be target-agnostic
  memory: Extract memory_region_msync() from memory_region_writeback()
  memory: Rename memory_region_do_writeback -> memory_region_writeback
  fuzz: run the main-loop in fork-server process
  fuzz: add mangled object name to linker script
  fuzz: fix typo in i440fx-qtest-reboot arguments
  fuzz: add datadir for oss-fuzz compatability
  io_uring: use io_uring_cq_ready() to check for ready cqes
  io_uring: retry io_uring_submit() if it fails with errno=EINTR

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

block: Factor out bdrv_run_co()

We have a few bdrv_*() functions that can either spawn a new coroutine
and wait for it with BDRV_POLL_WHILE() or use a fastpath if they are
alreeady running in a coroutine. All of them duplicate basically the
same code.

Factor the common code into a new function bdrv_run_co().

Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20200520144901.16589-1-vsementsov@virtuozzo.com
   [Factor out bdrv_run_co_entry too]
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

exec: Rename qemu_ram_writeback() as qemu_ram_msync()

Rename qemu_ram_writeback() as qemu_ram_msync() to better
match what it does.

Suggested-by: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20200508062456.23344-5-philmd@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

hw/block: Let the NVMe emulated device be target-agnostic

Now than the non-target specific memory_region_msync() function
is available, use it to make this device target-agnostic.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20200508062456.23344-4-philmd@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

memory: Extract memory_region_msync() from memory_region_writeback()

Suggested-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20200508062456.23344-3-philmd@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

memory: Rename memory_region_do_writeback -> memory_region_writeback

We usually use '_do_' for internal functions. Rename
memory_region_do_writeback() as memory_region_writeback().

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20200508062456.23344-2-philmd@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

fuzz: run the main-loop in fork-server process

Without this, the time since the last main-loop keeps increasing, as the
fuzzer runs. The forked children need to handle all the "past-due"
timers, slowing them down, over time. With this change, the
parent/fork-server process runs the main-loop, while waiting on the
child, ensuring that the timer events do not pile up, over time.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Message-id: 20200512030133.29896-5-alxndr@bu.edu
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

fuzz: add mangled object name to linker script

Previously, we relied on "FuzzerTracePC*(.bss*)" to place libfuzzer's
fuzzer::TPC object into our contiguous shared-memory region. This does
not work for some libfuzzer builds, so this addition identifies the
region by its mangled name: *(.bss._ZN6fuzzer3TPCE);

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Message-id: 20200512030133.29896-4-alxndr@bu.edu
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

fuzz: fix typo in i440fx-qtest-reboot arguments

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20200512030133.29896-3-alxndr@bu.edu
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

fuzz: add datadir for oss-fuzz compatability

This allows us to keep pc-bios in executable_dir/pc-bios, rather than
executable_dir/../pc-bios, which is incompatible with oss-fuzz' file
structure.

Signed-off-by: Alexander Bulekov <alxndr@bu.edu>
Reviewed-by: Darren Kenny <darren.kenny@oracle.com>
Message-id: 20200512030133.29896-2-alxndr@bu.edu
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

io_uring: use io_uring_cq_ready() to check for ready cqes

In qemu_luring_poll_cb() we are not using the cqe peeked from the
CQ ring. We are using io_uring_peek_cqe() only to see if there
are cqes ready, so we can replace it with io_uring_cq_ready().

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20200519134942.118178-1-sgarzare@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

io_uring: retry io_uring_submit() if it fails with errno=EINTR

As recently documented [1], io_uring_enter(2) syscall can return an
error (errno=EINTR) if the operation was interrupted by a delivery
of a signal before it could complete.

This should happen when IORING_ENTER_GETEVENTS flag is used, for
example during io_uring_submit_and_wait() or during io_uring_submit()
when IORING_SETUP_IOPOLL is enabled.

We shouldn't have this problem for now, but it's better to prevent it.

[1] https://github.com/axboe/liburing/commit/344355ec6619de8f4e64584c9736530b5346e4f4

Signed-off-by: Stefano Garzarella <sgarzare@redhat.com>
Message-id: 20200519133041.112138-1-sgarzare@redhat.com
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge remote-tracking branch 'remotes/alistair/tags/pull-riscv-to-apply-20200603' into staging

This is a collection of RISC-V patches for 5.1.

This incldues removing deprecated features and part of the OpenTitan
support series.

# gpg: Signature made Wed 03 Jun 2020 17:12:43 BST
# gpg:                using RSA key F6C4AC46D4934868D3B8CE8F21E10D29DF977054
# gpg: Good signature from "Alistair Francis <alistair@alistair23.me>" [full]
# Primary key fingerprint: F6C4 AC46 D493 4868 D3B8  CE8F 21E1 0D29 DF97 7054

* remotes/alistair/tags/pull-riscv-to-apply-20200603:
  riscv: Initial commit of OpenTitan machine
  target/riscv: Add the lowRISC Ibex CPU
  target/riscv: Don't set PMP feature in the cpu init
  target/riscv: Disable the MMU correctly
  target/riscv: Don't overwrite the reset vector
  riscv/boot: Add a missing header include
  riscv: sifive_e: Manually define the machine
  docs: deprecated: Update the -bios documentation
  target/riscv: Drop support for ISA spec version 1.09.1
  target/riscv: Remove the deprecated CPUs
  hw/riscv: spike: Remove deprecated ISA specific machines
  hw/riscv: virt: Remove the riscv_ prefix of the machine* functions
  hw/riscv: sifive_u: Remove the riscv_ prefix of the soc* functions
  riscv: Change the default behavior if no -bios option is specified
  riscv: Suppress the error report for QEMU testing with riscv_find_firmware()

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

riscv: Initial commit of OpenTitan machine

This adds a barebone OpenTitan machine to QEMU.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

target/riscv: Add the lowRISC Ibex CPU

Ibex is a small and efficient, 32-bit, in-order RISC-V core with
a 2-stage pipeline that implements the RV32IMC instruction set
architecture.

For more details on lowRISC see here:
https://github.com/lowRISC/ibex

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: LIU Zhiwei <zhiwei_liu@c-sky.com>

target/riscv: Don't set PMP feature in the cpu init

The PMP is enabled by default via the "pmp" property so there is no need
for us to set it in the init function. As all CPUs have PMP support just
remove the set_feature() call in the CPU init functions.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

target/riscv: Disable the MMU correctly

Previously if we didn't enable the MMU it would be enabled in the
realize() function anyway. Let's ensure that if we don't want the MMU we
disable it. We also don't need to enable the MMU as it will be enabled
in realize() by default.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

target/riscv: Don't overwrite the reset vector

The reset vector is set in the init function don't set it again in
realize.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

riscv/boot: Add a missing header include

As the functions declared in this header use the symbol_fn_t
typedef itself declared in "hw/loader.h", we need to include
it here to make the header file self-contained.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

riscv: sifive_e: Manually define the machine

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Palmer Dabbelt <palmerdabbelt@google.com>

docs: deprecated: Update the -bios documentation

Update the -bios deprecation documentation to describe the new
behaviour.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

target/riscv: Drop support for ISA spec version 1.09.1

The RISC-V ISA spec version 1.09.1 has been deprecated in QEMU since
4.1. It's not commonly used so let's remove support for it.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

target/riscv: Remove the deprecated CPUs

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>

hw/riscv: spike: Remove deprecated ISA specific machines

The ISA specific Spike machines have been deprecated in QEMU since 4.1,
let's finally remove them.

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>

hw/riscv: virt: Remove the riscv_ prefix of the machine* functions

Remove the riscv_ prefix of the machine* functions.

Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1590072147-13035-2-git-send-email-bmeng.cn@gmail.com
Message-Id: <1590072147-13035-2-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

hw/riscv: sifive_u: Remove the riscv_ prefix of the soc* functions

To keep consistency with the machine* functions, remove the riscv_
prefix of the soc* functions.

Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1590072147-13035-1-git-send-email-bmeng.cn@gmail.com
Message-Id: <1590072147-13035-1-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

riscv: Change the default behavior if no -bios option is specified

Per QEMU deprecated doc, QEMU 4.1 introduced support for the -bios
option in QEMU for RISC-V for the virt machine and sifive_u machine.
The default behavior has been that QEMU does not automatically load
any firmware if no -bios option is included.

Now 2 releases passed, it's time to change the default behavior to
load the default OpenSBI firmware automatically. The firmware is
included with the QEMU release and no user interaction is required.
All a user needs to do is specify the kernel they want to boot with
the -kernel option.

Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Message-id: 1588335545-649-1-git-send-email-bmeng.cn@gmail.com
Message-Id: <1588335545-649-1-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

riscv: Suppress the error report for QEMU testing with riscv_find_firmware()

We only ship plain binary bios images in the QEMU source. With Spike
machine that uses ELF images as the default bios, running QEMU test
will complain hence let's suppress the error report for QEMU testing.

Signed-off-by: Bin Meng <bin.meng@windriver.com>
Reviewed-by: Anup Patel <anup@brainfault.org>
Message-Id: <1588348254-7241-6-git-send-email-bmeng.cn@gmail.com>
Signed-off-by: Alistair Francis <alistair.francis@wdc.com>

Merge remote-tracking branch 'remotes/rth/tags/pull-tcg-20200602' into staging

Vector rotate support
Signal handling support for NetBSD arm/aarch64

# gpg: Signature made Tue 02 Jun 2020 17:43:05 BST
# gpg:                using RSA key 7A481E78868B4DB6A85A05C064DF38E8AF7E215F
# gpg:                issuer "richard.henderson@linaro.org"
# gpg: Good signature from "Richard Henderson <richard.henderson@linaro.org>" [full]
# Primary key fingerprint: 7A48 1E78 868B 4DB6 A85A  05C0 64DF 38E8 AF7E 215F

* remotes/rth/tags/pull-tcg-20200602:
  accel/tcg: Provide a NetBSD specific aarch64 cpu_signal_handler
  accel/tcg: Adjust cpu_signal_handler for NetBSD/arm
  tcg: Improve move ops in liveness_pass_2
  target/s390x: Use tcg_gen_gvec_rotl{i,s,v}
  target/ppc: Use tcg_gen_gvec_rotlv
  tcg/ppc: Implement INDEX_op_rot[lr]v_vec
  tcg/aarch64: Implement INDEX_op_rotl{i,v}_vec
  tcg/i386: Implement INDEX_op_rotl{i,s,v}_vec
  tcg: Implement gvec support for rotate by scalar
  tcg: Remove expansion to shift by vector from do_shifts
  tcg: Implement gvec support for rotate by vector
  tcg: Implement gvec support for rotate by immediate

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

accel/tcg: Provide a NetBSD specific aarch64 cpu_signal_handler

Fix qemu build on NetBSD/evbarm-aarch64 by providing a NetBSD specific
cpu_signal_handler.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Nick Hudson <skrll@netbsd.org>
Message-Id: <20200517101529.5367-1-skrll@netbsd.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

accel/tcg: Adjust cpu_signal_handler for NetBSD/arm

Fix building on NetBSD/arm by extracting the FSR value from the
correct siginfo_t field.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Nick Hudson <skrll@netbsd.org>
Message-Id: <20200516154147.24842-1-skrll@netbsd.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg: Improve move ops in liveness_pass_2

If the output of the move is dead, then the last use is in
the store.  If we propagate the input to the store, then we
can remove the move opcode entirely.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

target/s390x: Use tcg_gen_gvec_rotl{i,s,v}

Merge VERLL and VERLLV into op_vesv and op_ves, alongside
all of the other vector shift operations.

Reviewed-by: David Hildenbrand <david@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

target/ppc: Use tcg_gen_gvec_rotlv

Acked-by: David Gibson <david@gibson.dropbear.id.au>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg/ppc: Implement INDEX_op_rot[lr]v_vec

We already had support for rotlv, using a target-specific opcode;
convert to use the generic opcode.  Handle rotrv via simple negation.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg/aarch64: Implement INDEX_op_rotl{i,v}_vec

For immediate rotate , we can implement this in two instructions,
using SLI.  For variable rotate, the oddness of aarch64 right-shift-
as-negative-left-shift means a backend-specific expansion works best.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg/i386: Implement INDEX_op_rotl{i,s,v}_vec

For immediates, we must continue the special casing of 8-bit
elements.  The other element sizes and shift types are trivially
implemented with shifts.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg: Implement gvec support for rotate by scalar

No host backend support yet, but the interfaces for rotls
are in place.  Only implement left-rotate for now, as the
only known use of vector rotate by scalar is s390x, so any
right-rotate would be unused and untestable.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg: Remove expansion to shift by vector from do_shifts

We do not reflect this expansion in tcg_can_emit_vecop_list,
so it is unused and unusable.  However, we actually perform
the same expansion in do_gvec_shifts, so it is also unneeded.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

tcg: Implement gvec support for rotate by vector

No host backend support yet, but the interfaces for rotlv
and rotrv are in place.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
v3: Drop the generic expansion from rot to shift; we can do better
    for each backend, and then this code becomes unused.

tcg: Implement gvec support for rotate by immediate

No host backend support yet, but the interfaces for rotli
are in place.  Canonicalize immediate rotate to the left,
based on a survey of architectures, but provide both left
and right shift interfaces to the translators.

Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>

Merge remote-tracking branch 'remotes/vivier/tags/m68k-for-5.1-pull-request' into staging

Two fixes for the 68040 FPU

# gpg: Signature made Tue 02 Jun 2020 16:21:19 BST
# gpg:                using RSA key CD2F75DDC8E3A4DC2E4F5173F30C38BD3F2FBE3C
# gpg:                issuer "laurent@vivier.eu"
# 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/vivier/tags/m68k-for-5.1-pull-request:
  target/m68k: implement opcode fetoxm1
  target/m68k: implement fmove.l #<data>,FPCR

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

target/m68k: implement opcode fetoxm1

Example provided in the launchpad bug fails with:

   qemu: uncaught target signal 4 (Illegal instruction) - core dumped
   Illegal instruction (core dumped)

It appears fetoxm1 is not implemented:

   IN: expm1f
   0x800005cc: fetoxm1x %fp2,%fp0
   Disassembler disagrees with translator over instruction decoding
   Please report this to qemu-devel@nongnu.org

   (gdb) x/2hx 0x800005cc
   0x800005cc: 0xf200 0x0808

This patch adds the instruction.

Bug: https://bugs.launchpad.net/qemu/+bug/1881450
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200531131951.631902-1-laurent@vivier.eu>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

target/m68k: implement fmove.l #<data>,FPCR

The immediate value mode was ignored and instruction execution
ends to an invalid access mode.

This was found running 'R' that set FPSR to 0 at startup with
a 'fmove.l #0,FPSR' in qemu-system-m68k emulation and triggers a
kernel crash:

[   56.640000] *** ADDRESS ERROR ***   FORMAT=2
[   56.640000] Current process id is 728
[   56.640000] BAD KERNEL TRAP: 00000000
[   56.640000] Modules linked in: sg evdev mac_hid ip_tables x_tables sha1_generic hmac ipv6 nf_defrag_ipv6 autofs4 ext4 crc16 mbcache jbd2 crc32c_generic sd_mod t10_pi crc_t10dif crct10dif_generic crct10dif_common sr_mod cdrom mac_esp macsonic esp_scsi
[   56.640000] PC: [<00016a2c>] X_UNSUPP+0x2c/0x3c
[   56.640000] SR: 2004  SP: 3eb5e68c  a2: c02e239a
[   56.640000] d0: 00000040    d1: 00000002    d2: 8002adec    d3: 8002ad50
[   56.640000] d4: 8002c768    d5: 0000000d    a0: ffffffc2    a1: ffffffc1
[   56.640000] Process R (pid: 728, task=a3dfda5d)
[   56.640000] Frame format=2 instr addr=00000000
[   56.650000] Stack from 3a4d9f30:
[   56.650000]         41000000 00000002 00000002 ffffffc2 ffffffc1 1fff0000 80000000 00000000
[   56.650000]         3fbf0000 80000000 00000000 00000000 20000000 00000000 7fff0000 ffffffff
[   56.650000]         ffffffff 00000000 00050008 00000000 8000067c c02c2000 efffee20 000002d8
[   56.650000]         00002a28 3a4d9f98 00000002 00000014 fffffffe 8002c768 00000002 00000041
[   56.650000]         00000002 c041fc58 c0743758 ffffffff 00000000 0008c075 00002b24 00000012
[   56.650000]         000007d0 00000024 00000002 c05bef04 c05bef04 0000005e 00000077 c28aca70
[   56.650000] Call Trace: [<00050008>] copy_overflow+0x10/0x28
[   56.650000]  [<00002a28>] buserr+0x20/0x28
[   56.650000]  [<0008c075>] bpf_check+0x57f/0x1cfa
[   56.650000]  [<00002b24>] syscall+0x8/0xc
[   56.650000]  [<0000c019>] dn_sched_init+0x75/0x88
[   56.650000] Code: 1017 0200 00f0 0c00 0040 66ff 0000 05ac <f23c> 8800 0000 0000 f23c 9000 0000 0000 222e ff84 082e 0005 ff1c 6600 000a 0281
[   56.650000] Disabling lock debugging due to kernel taint
...

Reported-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20200531110231.620711-1-laurent@vivier.eu>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>

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

machine queue, 2020-05-13

Bug fixes:
* hostmem: don't use mbind() if host-nodes is empty (Igor Mammedov)

# gpg: Signature made Wed 13 May 2020 15:00:25 BST
# gpg:                using RSA key 5A322FD5ABC4D3DBACCFD1AA2807936F984DC5A6
# gpg:                issuer "ehabkost@redhat.com"
# 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/machine-next-pull-request:
  hostmem: don't use mbind() if host-nodes is empty

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

Merge remote-tracking branch 'remotes/dgilbert/tags/pull-migration-20200601a' into staging

Migration/virtio/hmp pull 2020-06-01

A mixed pull with:
  - RDMA migration fix (CID 1428762)
  - HMP qom-get addition and qom-set cleanup
  - a virtiofsd fix
  - COLO fixes

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
# gpg: Signature made Mon 01 Jun 2020 19:37:15 BST
# gpg:                using RSA key 45F5C71B4A0CB7FB977A9FA90516331EBC5BFDE7
# gpg: Good signature from "Dr. David Alan Gilbert (RH2) <dgilbert@redhat.com>" [full]
# Primary key fingerprint: 45F5 C71B 4A0C B7FB 977A  9FA9 0516 331E BC5B FDE7

* remotes/dgilbert/tags/pull-migration-20200601a:
  migration/migration.c: Fix hang in ram_save_host_page
  migration/colo.c: Move colo_notify_compares_event to the right place
  migration/colo.c: Relaunch failover even if there was an error
  migration/colo.c: Flush ram cache only after receiving device state
  migration/colo.c: Use cpu_synchronize_all_states()
  migration/colo.c: Use event instead of semaphore
  migration/vmstate: Remove unnecessary MemoryRegion forward declaration
  virtiofsd: remove symlink fallbacks
  hmp: Simplify qom-set
  hmp: Implement qom-get HMP command
  migration/rdma: cleanup rdma context before g_free to avoid memleaks
  migration/rdma: fix potential nullptr access in rdma_start_incoming_migration

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

Merge remote-tracking branch 'remotes/bkoppelmann2/tags/pull-tricore-20200601' into staging

Remove ctx->env ptr, add TriCore gdb stub

# gpg: Signature made Mon 01 Jun 2020 16:06:35 BST
# gpg:                using RSA key 6E636A7E83F2DD0CFA6E6E370AD2C6396B69CA14
# gpg:                issuer "kbastian@mail.uni-paderborn.de"
# gpg: Good signature from "Bastian Koppelmann <kbastian@mail.uni-paderborn.de>" [full]
# Primary key fingerprint: 6E63 6A7E 83F2 DD0C FA6E  6E37 0AD2 C639 6B69 CA14

* remotes/bkoppelmann2/tags/pull-tricore-20200601:
  target/tricore: Implement gdbstub
  target/tricore: Implement tricore_cpu_get_phys_page_debug
  target/tricore: Raise EXCP_DEBUG in gen_goto_tb() for singlestep
  target/tricore: Move translate feature check to ctx
  target/tricore: Don't save pc in generate_qemu_excp

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

migration/migration.c: Fix hang in ram_save_host_page

migration_rate_limit will erroneously ratelimit a shutdown socket,
which causes the migration thread to hang in ram_save_host_page
if the socket is shutdown.

Fix this by explicitly testing if the socket has errors or was
shutdown in migration_rate_limit.

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
Message-Id: <e79085bbe2d46dfa007dd41820194d5e2d4fcd80.1590007004.git.lukasstraub2@web.de>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/colo.c: Move colo_notify_compares_event to the right place

If the secondary has to failover during checkpointing, it still is
in the old state (i.e. different state than primary). Thus we can't
expose the primary state until after the checkpoint is sent.

This fixes sporadic connection reset of client connections during
failover.

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
Message-Id: <d4555dd5146a54518c4d9d4efd996b7c745c6687.1589193382.git.lukasstraub2@web.de>
Reviewed-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/colo.c: Relaunch failover even if there was an error

If vmstate_loading is true, secondary_vm_do_failover will set failover
status to FAILOVER_STATUS_RELAUNCH and return success without initiating
failover. However, if there is an error during the vmstate_loading
section, failover isn't relaunched. Instead we then wait for
failover on colo_incoming_sem.

Fix this by relaunching failover even if there was an error. Also,
to make this work properly, set vmstate_loading to false when
returning during the vmstate_loading section.

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
Message-Id: <f60b0a8e2fadaaec792e04819dfc46951842d6ba.1589193382.git.lukasstraub2@web.de>
Reviewed-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/colo.c: Flush ram cache only after receiving device state

If we suceed in receiving ram state, but fail receiving the device
state, there will be a mismatch between the two.

Fix this by flushing the ram cache only after the vmstate has been
received.

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
Message-Id: <3289d007d494cb0e2f05b1cf4ae6a78d300fede3.1589193382.git.lukasstraub2@web.de>
Reviewed-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/colo.c: Use cpu_synchronize_all_states()

cpu_synchronize_all_pre_loadvm() marks all vcpus as dirty, so the
registers are loaded from CPUState before we continue running
the vm. However if we failover during checkpoint, CPUState is not
initialized and the registers are loaded with garbage. This causes
guest hangs and crashes.

Fix this by using cpu_synchronize_all_states(), which initializes
CPUState from the current cpu registers additionally to marking
the vcpus as dirty.

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
Message-Id: <9675031ce557b73ebd10e7bd20ebbf57f30b177c.1589193382.git.lukasstraub2@web.de>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/colo.c: Use event instead of semaphore

If multiple packets miscompare in a short timeframe, the semaphore
value will be increased multiple times. This causes multiple
checkpoints even if one would be sufficient.

Fix this by using a event instead of a semaphore for triggering
checkpoints. Now, checkpoint requests will be ignored until the
checkpoint event is sent to colo-compare (which releases the
miscompared packets).

Benchmark results (iperf3):
Client-to-server tcp:
without patch: ~66 Mbit/s
with patch: ~61 Mbit/s
Server-to-client tcp:
without patch: ~702 Kbit/s
with patch: ~16 Mbit/s

Signed-off-by: Lukas Straub <lukasstraub2@web.de>
Message-Id: <fd601ba1beb524aada54ba66e87ebfc12cf4574b.1589193382.git.lukasstraub2@web.de>
Reviewed-by: zhanghailiang <zhang.zhanghailiang@huawei.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/vmstate: Remove unnecessary MemoryRegion forward declaration

"migration/vmstate.h" only uses pointer to MemoryRegion, which
is already forward declared in "qemu/typedefs.h".

Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20200530165512.15225-1-f4bug@amsat.org>
Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

virtiofsd: remove symlink fallbacks

Path lookup in the kernel has special rules for looking up magic symlinks
under /proc.  If a filesystem operation is instructed to follow symlinks
(e.g. via AT_SYMLINK_FOLLOW or lack of AT_SYMLINK_NOFOLLOW), and the final
component is such a proc symlink, then the target of the magic symlink is
used for the operation, even if the target itself is a symlink.  I.e. path
lookup is always terminated after following a final magic symlink.

I was erronously assuming that in the above case the target symlink would
also be followed, and so workarounds were added for a couple of operations
to handle the symlink case.  Since the symlink can be handled simply by
following the proc symlink, these workardouds are not needed.

Also remove the "norace" option, which disabled the workarounds.

Commit bdfd66788349 ("virtiofsd: Fix xattr operations") already dealt with
the same issue for xattr operations.

Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Message-Id: <20200514140736.20561-1-mszeredi@redhat.com>
Acked-by: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

hmp: Simplify qom-set

Simplify qom_set by making it use qmp_qom_set and the JSON parser.

(qemu) qom-get /machine smm
"auto"
(qemu) qom-set /machine smm "auto"

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200520151108.160598-3-dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
  With 's'->'S' type change suggested by Paolo and Markus

hmp: Implement qom-get HMP command

This started off as Andreas Färber's implementation from
March 2015, but after feedback from Paolo and Markus it morphed into
using the json output which handles structs reasonably.

Use with qom-list to find the members of an object.

(qemu) qom-get /backend/console[0]/device/vga.rom[0] size
65536
(qemu) qom-get /machine smm
"auto"
(qemu) qom-get /machine rtc-time
{
    "tm_year": 120,
    "tm_sec": 51,
    "tm_hour": 9,
    "tm_min": 50,
    "tm_mon": 4,
    "tm_mday": 20
}
(qemu) qom-get /machine frob
Error: Property '.frob' not found

Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20200520151108.160598-2-dgilbert@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/rdma: cleanup rdma context before g_free to avoid memleaks

When error happen in initializing 'rdma_return_path', we should cleanup rdma context
before g_free(rdma) to avoid some memleaks. This patch fix that.

Reported-by: Euler Robot <euler.robot@huawei.com>
Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
Message-Id: <20200508100755.7875-3-pannengyuan@huawei.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>

migration/rdma: fix potential nullptr access in rdma_start_incoming_migration

'rdma' is NULL when taking the first error branch in rdma_start_incoming_migration.
And it will cause a null pointer access in label 'err'. Fix that.

Fixes: 59c59c67ee6b0327ae932deb303caa47919aeb1e
Signed-off-by: Pan Nengyuan <pannengyuan@huawei.com>
Message-Id: <20200508100755.7875-2-pannengyuan@huawei.com>
Reviewed-by: Juan Quintela <quintela@redhat.com>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
  Note this is CID 1428762

target/tricore: Implement gdbstub

Acked-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Message-Id: <20200529072148.284037-6-kbastian@mail.uni-paderborn.de>

target/tricore: Implement tricore_cpu_get_phys_page_debug

this also removes tricore_cpu_get_phys_page_attrs_debug() as it was a
temporary fix from b190f477e29c7cd03a8fee49c96d27f160e3f5b0.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Message-Id: <20200529072148.284037-5-kbastian@mail.uni-paderborn.de>

target/tricore: Raise EXCP_DEBUG in gen_goto_tb() for singlestep

this is needed for remote gdb connections.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Message-Id: <20200529072148.284037-4-kbastian@mail.uni-paderborn.de>

target/tricore: Move translate feature check to ctx

this allows us to remove the references to env from ctx. This also fixes
a segfault that was due to the unititalized ctx->env ptr.

Reported-by: Andreas Konopik <andreas.konopik@fau.de>
Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Message-Id: <20200529072148.284037-3-kbastian@mail.uni-paderborn.de>

target/tricore: Don't save pc in generate_qemu_excp

EXCP_DEBUG is the only user. If we encounter a jump in tricore-gdb it's
target was overwritten by generate_qemu_excp() and we would never leave.

Signed-off-by: Bastian Koppelmann <kbastian@mail.uni-paderborn.de>
Message-Id: <20200529072148.284037-2-kbastian@mail.uni-paderborn.de>

Merge remote-tracking branch 'remotes/amarkovic/tags/mips-queue-june-01-2020' into staging

MIPS queue for June 1st, 2020

# gpg: Signature made Mon 01 Jun 2020 12:29:25 BST
# gpg:                using RSA key D4972A8967F75A65
# gpg: Good signature from "Aleksandar Markovic <amarkovic@wavecomp.com>" [full]
# Primary key fingerprint: 8526 FBF1 5DA3 811F 4A01  DD75 D497 2A89 67F7 5A65

* remotes/amarkovic/tags/mips-queue-june-01-2020:
  hw/mips: fuloong2e: Set preferred page size to 16KB
  target/mips: Support variable page size
  target/mips: Add more CP0 register for save/restore
  hw/mips: Add CPU IRQ3 delivery for KVM
  configure: Add KVM target support for MIPS64
  tests/Makefile: Fix description of "make check"

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

hw/mips: fuloong2e: Set preferred page size to 16KB

Loongson processor prefers 16KB page size in system emulator, so let's
define mc->minimum_page_bits to 14.

Cc: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhc@lemote.com>
Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Signed-off-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Message-Id: <1586337380-25217-2-git-send-email-chenhc@lemote.com>

target/mips: Support variable page size

Traditionally, MIPS use 4KB page size, but Loongson prefer 16KB page
size in system emulator. So, let's define TARGET_PAGE_BITS_VARY and
TARGET_PAGE_BITS_MIN to support variable page size.

Cc: Jiaxun Yang <jiaxun.yang@flygoat.com>
Signed-off-by: Huacai Chen <chenhc@lemote.com>
Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Signed-off-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Message-Id: <1586337380-25217-1-git-send-email-chenhc@lemote.com>

target/mips: Add more CP0 register for save/restore

Add more CP0 register for save/restore, including: EBase, XContext,
PageGrain, PWBase, PWSize, PWField, PWCtl, Config*, KScratch1~KScratch6.

Signed-off-by: Huacai Chen <chenhc@lemote.com>
Co-developed-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Signed-off-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Message-Id: <1588501221-1205-6-git-send-email-chenhc@lemote.com>

hw/mips: Add CPU IRQ3 delivery for KVM

Currently, KVM/MIPS only deliver I/O interrupt via IP2, this patch add
IP3 delivery as well, because Loongson-3 based machine use both IRQ2
(CPU's IP2) and IRQ3 (CPU's IP3).

Signed-off-by: Huacai Chen <chenhc@lemote.com>
Co-developed-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Signed-off-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Message-Id: <1588501221-1205-4-git-send-email-chenhc@lemote.com>

configure: Add KVM target support for MIPS64

Preparing for Loongson-3 virtualization, add KVM target support for
MIPS64 in configure script.

Signed-off-by: Huacai Chen <chenhc@lemote.com>
Co-developed-by: Jiaxun Yang <jiaxun.yang@flygoat.com>
Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Signed-off-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Message-Id: <1588501221-1205-2-git-send-email-chenhc@lemote.com>

tests/Makefile: Fix description of "make check"

The description of "make check" is out-of-date, so fix it by adding
block and softfloat.

Reviewed-by: Claudio Fontana <cfontana@suse.de>
Signed-off-by: Huacai Chen <chenhc@lemote.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Signed-off-by: Aleksandar Markovic <aleksandar.qemu.devel@gmail.com>
Message-Id: <1588674291-6486-1-git-send-email-chenhc@lemote.com>

Merge remote-tracking branch 'remotes/philmd-gitlab/tags/python-next-20200531' into staging

Python queue:

* migration acceptance test fix
* introduce pylintrc & flake8 config
* various cleanups (Python3, style)
* vm-test can set QEMU_LOCAL=1 to use locally built binaries
* refactored BootLinuxBase & LinuxKernelTest acceptance classes

https://gitlab.com/philmd/qemu/pipelines/151323210
https://travis-ci.org/github/philmd/qemu/builds/693157969

# gpg: Signature made Sun 31 May 2020 17:37:35 BST
# gpg:                using RSA key FAABE75E12917221DCFD6BB2E3E32C2CDEADC0DE
# gpg: Good signature from "Philippe Mathieu-Daudé (F4BUG) <f4bug@amsat.org>" [full]
# Primary key fingerprint: FAAB E75E 1291 7221 DCFD  6BB2 E3E3 2C2C DEAD C0DE

* remotes/philmd-gitlab/tags/python-next-20200531: (25 commits)
  tests/acceptance: refactor boot_linux to allow code reuse
  tests/acceptance: refactor boot_linux_console test to allow code reuse
  tests/acceptance: allow console interaction with specific VMs
  tests/acceptance/migration.py: Wait for both sides
  tests/migration/guestperf: Use Python 3 interpreter
  tests/vm: allow wait_ssh() to specify command
  tests/vm: Add ability to select QEMU from current build
  tests/vm: Pass --debug through for vm-boot-ssh
  python/qemu/qtest: Check before accessing _qtest
  python/qemu/qmp: assert sockfile is not None
  python/qemu/qmp: use True/False for non/blocking modes
  python/qemu: Adjust traceback typing
  python/qemu: fix socket.makefile() typing
  python/qemu: remove Python2 style super() calls
  python/qemu: delint; add flake8 config
  python/qemu: delint and add pylintrc
  python/qemu/machine: remove logging configuration
  python/qemu/machine: add kill() method
  python: remove more instances of sys.version_info
  scripts/qmp: Fix shebang and imports
  ...

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

Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2020-05-30' into staging

- Replace hw_error() with qemu_log_mask() in the m68k coldfire machine code

# gpg: Signature made Sat 30 May 2020 08:44:41 BST
# gpg:                using RSA key 27B88847EEE0250118F3EAB92ED9D774FE702DB5
# gpg:                issuer "huth@tuxfamily.org"
# 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-2020-05-30:
  hw/m68k/mcf52xx: Replace hw_error() by qemu_log_mask()
  hw/m68k/mcf5206: Reduce m5206_mbar_read/write() offset arg to 16-bit

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

tests/acceptance: refactor boot_linux to allow code reuse

This patch moves image downloading functions to the separate class to allow
reusing them from record/replay tests.

Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <159073593167.20809.17582679291556188984.stgit@pasha-ThinkPad-X280>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

tests/acceptance: refactor boot_linux_console test to allow code reuse

This patch splits code in BootLinuxConsole class into two different
classes to allow reusing it by record/replay tests.

Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <159073588490.20809.13942096070255577558.stgit@pasha-ThinkPad-X280>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

tests/acceptance: allow console interaction with specific VMs

Console interaction in avocado scripts was possible only with single
default VM.
This patch modifies the function parameters to allow passing a specific
VM as a parameter to interact with it.

Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Reviewed-by: Willian Rampazzo <willianr@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <159073587933.20809.5122618715976660635.stgit@pasha-ThinkPad-X280>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

tests/acceptance/migration.py: Wait for both sides

When the source finishes migration the destination will still be
receiving the data sent by the source, so it might not have quite
finished yet, so won't quite have reached 'completed'.
This lead to occasional asserts in the next few checks.

After the source has finished, check the destination as well.
(We can't just switch to checking the destination, because it doesn't
give a status until it has started receiving the migration).

Reported-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200528112404.121972-1-dgilbert@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

tests/migration/guestperf: Use Python 3 interpreter

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Kevin Wolf <kwolf@redhat.com>
Message-Id: <20200512103238.7078-7-philmd@redhat.com>

tests/vm: allow wait_ssh() to specify command

This allows for waiting for completion of arbitrary commands.

Signed-off-by: Robert Foley <robert.foley@linaro.org>
Reviewed-by: Peter Puhov <peter.puhov@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200529203458.1038-7-robert.foley@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

tests/vm: Add ability to select QEMU from current build

Added a new special variable QEMU_LOCAL=1, which
will indicate to take the QEMU binary from the current
build.

Signed-off-by: Robert Foley <robert.foley@linaro.org>
Reviewed-by: Peter Puhov <peter.puhov@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200529203458.1038-6-robert.foley@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

tests/vm: Pass --debug through for vm-boot-ssh

This helps debug issues that occur during the boot sequence.

Signed-off-by: Robert Foley <robert.foley@linaro.org>
Reviewed-by: Peter Puhov <peter.puhov@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200529203458.1038-5-robert.foley@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu/qtest: Check before accessing _qtest

It can be None; so add assertions or exceptions where appropriate to
guard the access accordingly.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-30-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu/qmp: assert sockfile is not None

In truth, if you don't do this, you'll just get a TypeError
exception. Now, you'll get an AssertionError.

Is this tangibly better? No.
Does mypy complain less? Yes.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-21-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu/qmp: use True/False for non/blocking modes

The type system doesn't want integers.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-15-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu: Adjust traceback typing

mypy considers it incorrect to use `bool` to statically return false,
because it will assume that it could conceivably return True, and gives
different analysis in that case. Use a None return to achieve the same
effect, but make mypy happy.

Note: Pylint considers function signatures as code that might trip the
duplicate-code checker. I'd rather not disable this as it does not
trigger often in practice, so I'm disabling it as a one-off and filed a
change request; see https://github.com/PyCQA/pylint/issues/3619

Signed-off-by: John Snow <jsnow@redhat.com>
Acked-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-14-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu: fix socket.makefile() typing

Note:

A bug in typeshed (https://github.com/python/typeshed/issues/3977)
misinterprets the type of makefile(). Work around this by explicitly
stating that we are opening a text-mode file.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-13-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu: remove Python2 style super() calls

Use the Python3 style instead.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-12-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu: delint; add flake8 config

Mostly, ignore the "no bare except" rule, because flake8 is not
contextual and cannot determine if we re-raise. Pylint can, though, so
always prefer pylint for that.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200528222129.23826-5-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu: delint and add pylintrc

Bring our these files up to speed with pylint 2.5.0.
Add a pylintrc file to formalize which pylint subset
we are targeting.

The similarity ignore is there to suppress similarity
reports across imports, which for typing constants,
are going to trigger this report erroneously.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200528222129.23826-4-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>

python/qemu/machine: remove logging configuration

Python 3.5 and above do not print a warning when logging is not
configured. As a library, it's best practice to leave logging
configuration to the client executable.

Signed-off-by: John Snow <jsnow@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200514055403.18902-22-jsnow@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>