qemu.git
2 years agohw/arm: Express dependencies of the virt machine with Kconfig
Thomas Huth [Fri, 8 Feb 2019 06:17:55 +0000 (07:17 +0100)] 
hw/arm: Express dependencies of the virt machine with Kconfig

Dependencies have been determined by looking at hw/arm/virt.c

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of the aspeed boards with Kconfig
Thomas Huth [Fri, 8 Feb 2019 05:53:04 +0000 (06:53 +0100)] 
hw/arm: Express dependencies of the aspeed boards with Kconfig

Dependencies have been determined by looking at hw/arm/aspeed.c

Reviewed-by: Cédric Le Goater <clg@kaod.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of collie with Kconfig
Thomas Huth [Fri, 8 Feb 2019 05:39:48 +0000 (06:39 +0100)] 
hw/arm: Express dependencies of collie with Kconfig

Add Kconfig dependencies for the Strongarm collie machine.
This patch is based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of xilinx-zynq with Kconfig
Thomas Huth [Thu, 7 Feb 2019 19:26:34 +0000 (20:26 +0100)] 
hw/arm: Express dependencies of xilinx-zynq with Kconfig

Add Kconfig dependencies for the xilinx-zynq-a9 board.
This patch is based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of the PXA2xx machines with Kconfig
Thomas Huth [Fri, 8 Feb 2019 11:35:26 +0000 (12:35 +0100)] 
hw/arm: Express dependencies of the PXA2xx machines with Kconfig

Add Kconfig dependencies for the PXA2xx machines (akita, borzoi,
connex and verdex gumstix, tosa, mainstone, spitz, terrier and z2).
This patch is based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of realview, versatile and vexpress with Kconfig
Thomas Huth [Thu, 7 Feb 2019 19:05:05 +0000 (20:05 +0100)] 
hw/arm: Express dependencies of realview, versatile and vexpress with Kconfig

This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of stellaris with Kconfig
Thomas Huth [Thu, 7 Feb 2019 16:33:50 +0000 (17:33 +0100)] 
hw/arm: Express dependencies of stellaris with Kconfig

This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of the OMAP machines with Kconfig
Thomas Huth [Thu, 7 Feb 2019 16:12:14 +0000 (17:12 +0100)] 
hw/arm: Express dependencies of the OMAP machines with Kconfig

Add Kconfig dependencies for the OMAP machines (cheetah, n800, n810,
sx1 and sx1-v1).
This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of musicpal with Kconfig
Thomas Huth [Thu, 7 Feb 2019 16:05:38 +0000 (17:05 +0100)] 
hw/arm: Express dependencies of musicpal with Kconfig

This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of the fsl-imx31 machine with Kconfig
Thomas Huth [Thu, 7 Feb 2019 15:51:56 +0000 (16:51 +0100)] 
hw/arm: Express dependencies of the fsl-imx31 machine with Kconfig

Add Kconfig dependencies for the fsl-imx31 / kzm machine.
This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Peter Chubb <peter.chubb@data61.csiro.au>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of integratorcp with Kconfig
Thomas Huth [Thu, 7 Feb 2019 14:21:27 +0000 (15:21 +0100)] 
hw/arm: Express dependencies of integratorcp with Kconfig

This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of the highbank machines with Kconfig
Thomas Huth [Thu, 7 Feb 2019 09:35:22 +0000 (10:35 +0100)] 
hw/arm: Express dependencies of the highbank machines with Kconfig

Add Kconfig dependencies for the highbank machine (and the midway
machine).
This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/arm: Express dependencies of the exynos machines with Kconfig
Thomas Huth [Thu, 7 Feb 2019 09:28:01 +0000 (10:28 +0100)] 
hw/arm: Express dependencies of the exynos machines with Kconfig

Add Kconfig dependencies for the Exynos-related boards (nuri and
smdkc210).
This patch is slightly based on earlier work by Ákos Kovács (i.e.
his "hw/arm/Kconfig: Add ARM Kconfig" patch).

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/ide/ahci: Add a Kconfig switch for the AHCI-ICH9 device
Thomas Huth [Fri, 22 Feb 2019 14:48:11 +0000 (15:48 +0100)] 
hw/ide/ahci: Add a Kconfig switch for the AHCI-ICH9 device

Some of our machines (like the ARM cubieboard) use CONFIG_AHCI for an AHCI
sysbus device, but do not use CONFIG_PCI since they do not feature a PCI
bus. With CONFIG_AHCI but without CONFIG_PCI, currently linking fails:

    ../hw/ide/ich.o: In function `pci_ich9_ahci_realize':
    hw/ide/ich.c:124: undefined reference to `pci_allocate_irq'
    hw/ide/ich.c:126: undefined reference to `pci_register_bar'
    hw/ide/ich.c:128: undefined reference to `pci_register_bar'
    hw/ide/ich.c:131: undefined reference to `pci_add_capability'
    hw/ide/ich.c:147: undefined reference to `msi_init'
    ../hw/ide/ich.o: In function `pci_ich9_uninit':
    hw/ide/ich.c:158: undefined reference to `msi_uninit'
    ../hw/ide/ich.o:(.data.rel+0x50): undefined reference to `vmstate_pci_device'

We must only compile ich.c if CONFIG_PCI is available, too, so introduce a
new config switch for this device.

Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: John Snow <jsnow@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agoMakefile: Fix inclusion of the config-devices.mak.d Kconfig dependencies
Thomas Huth [Fri, 10 May 2019 08:11:59 +0000 (10:11 +0200)] 
Makefile: Fix inclusion of the config-devices.mak.d Kconfig dependencies

The Makefile tries to include device Kconfig dependencies via

 -include $(SUBDIR_DEVICES_MAK_DEP)

and thus expects files that match *-softmmu/config-devices.mak.d ...
however, the minikconf script currently generates files a la
"*-softmmu-config.devices.mak.d" instead, so the dependency files
simply got ignored so far. For example, after a "touch hw/arm/Kconfig",
the arm-softmmu/config-devices.mak file is currently not re-generated.
Fix it by putting the dependency files in the *-softmmu folders now.

Reported-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/pci/pci-stub: Add msi_enabled() and msi_notify() to the pci stubs
Thomas Huth [Tue, 19 Feb 2019 16:00:10 +0000 (17:00 +0100)] 
hw/pci/pci-stub: Add msi_enabled() and msi_notify() to the pci stubs

Some machines have an AHCI adapter, but no PCI. To be able to
compile hw/ide/ahci.c without CONFIG_PCI, we still need the two
functions msi_enabled() and msi_notify() for linking.
This is required for the new Kconfig-like build system, if a user
wants to compile a QEMU binary with just one machine that has AHCI,
but no PCI, like the ARM "cubieboard" for example.

Reviewed-by: Michael S. Tsirkin <mst@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agoMerge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging
Peter Maydell [Fri, 10 May 2019 13:49:36 +0000 (14:49 +0100)] 
Merge remote-tracking branch 'remotes/stefanha/tags/block-pull-request' into staging

Pull request

# gpg: Signature made Fri 10 May 2019 14:02:22 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:
  docs: add Security chapter to the documentation
  docs: add Secure Coding Practices to developer docs
  aio-posix: ensure poll mode is left when aio_notify is called
  block/io.c: fix for the allocation failure
  block: Add coroutine_fn to bdrv_check_co_entry
  util/readline: Add braces to fix checkpatch errors
  util: readline: replace tab indent by four spaces to fix checkpatch errors
  util/readline: add a space to fix errors by checkpatch tool

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/vivier2/tags/linux-user-for-4.1-pull-request...
Peter Maydell [Fri, 10 May 2019 11:49:13 +0000 (12:49 +0100)] 
Merge remote-tracking branch 'remotes/vivier2/tags/linux-user-for-4.1-pull-request' into staging

GPROF fixes, GCC9 fixes, SIOCGIFNAME fix, new IPV6 sockopts, elf fix

# gpg: Signature made Fri 10 May 2019 11:45:23 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/linux-user-for-4.1-pull-request:
  linux-user: fix GPROF build failure
  linux-user: avoid treading on gprof's SIGPROF signals
  linux-user: elf: Map empty PT_LOAD segments
  The ioctl(SIOCGIFNAME) call requires a struct ifreq.
  linux-user: avoid string truncation warnings in uname field copying
  linux-user/elfload: Fix GCC 9 build warnings
  linux-user: Add missing IPV6 sockopts

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agolinux-user: fix GPROF build failure
Alex Bennée [Thu, 2 May 2019 09:27:28 +0000 (10:27 +0100)] 
linux-user: fix GPROF build failure

When linux-user/exit was introduced we failed to move the gprof
include at the same time. The CI didn't notice because it only builds
system emulation. Fix it for those that still find gprof useful.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Laurent Desnogues <laurent.desnogues@gmail.com>
Message-Id: <20190502092728.32727-1-alex.bennee@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agolinux-user: avoid treading on gprof's SIGPROF signals
Alex Bennée [Thu, 2 May 2019 14:58:46 +0000 (15:58 +0100)] 
linux-user: avoid treading on gprof's SIGPROF signals

The guest tends to get confused when it receives signals it doesn't
know about. Given the gprof magic has also set up it's own handler we
would do well to avoid stomping on it as well.

Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20190502145846.26226-1-alex.bennee@linaro.org>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agolinux-user: elf: Map empty PT_LOAD segments
Giuseppe Musacchio [Fri, 3 May 2019 12:20:07 +0000 (14:20 +0200)] 
linux-user: elf: Map empty PT_LOAD segments

Some PT_LOAD segments may be completely zeroed out and their p_filesize
is zero, in that case the loader should just allocate a page that's at
least p_memsz bytes large (plus eventual alignment padding).

Calling zero_bss does this job for us, all we have to do is make sure we
don't try to mmap a zero-length page.

Signed-off-by: Giuseppe Musacchio <thatlemon@gmail.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20190503122007.lkjsvztgt4ycovac@debian>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agoThe ioctl(SIOCGIFNAME) call requires a struct ifreq.
Erik Kline [Tue, 23 Apr 2019 22:20:05 +0000 (15:20 -0700)] 
The ioctl(SIOCGIFNAME) call requires a struct ifreq.

Signed-off-by: Erik Kline <ek@google.com>
Buglink: https://bugs.launchpad.net/qemu/+bug/1814352
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20190423222005.246981-1-ek@google.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agodocs: add Security chapter to the documentation
Stefan Hajnoczi [Thu, 9 May 2019 12:18:20 +0000 (13:18 +0100)] 
docs: add Security chapter to the documentation

This new chapter in the QEMU documentation covers the security
requirements that QEMU is designed to meet and principles for securely
deploying QEMU.

It is just a starting point that can be extended in the future with more
information.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-id: 20190509121820.16294-3-stefanha@redhat.com
Message-Id: <20190509121820.16294-3-stefanha@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agodocs: add Secure Coding Practices to developer docs
Stefan Hajnoczi [Thu, 9 May 2019 12:18:19 +0000 (13:18 +0100)] 
docs: add Secure Coding Practices to developer docs

At KVM Forum 2018 I gave a presentation on security in QEMU:
https://www.youtube.com/watch?v=YAdRf_hwxU8 (video)
https://vmsplice.net/~stefan/stefanha-kvm-forum-2018.pdf (slides)

This patch adds a guide to secure coding practices.  This document
covers things that developers should know about security in QEMU.  It is
just a starting point that we can expand on later.  I hope it will be
useful as a resource for new contributors and will save code reviewers
from explaining the same concepts many times.

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Acked-by: Stefano Garzarella <sgarzare@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Li Qiang <liq3ea@gmail.com>
Message-id: 20190509121820.16294-2-stefanha@redhat.com
Message-Id: <20190509121820.16294-2-stefanha@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoaio-posix: ensure poll mode is left when aio_notify is called
Paolo Bonzini [Tue, 9 Apr 2019 12:28:23 +0000 (14:28 +0200)] 
aio-posix: ensure poll mode is left when aio_notify is called

With aio=thread, adaptive polling makes latency worse rather than
better, because it delays the execution of the ThreadPool's
completion bottom half.

event_notifier_poll() does run while polling, detecting that
a bottom half was scheduled by a worker thread, but because
ctx->notifier is explicitly ignored in run_poll_handlers_once(),
scheduling the BH does not count as making progress and
run_poll_handlers() keeps running.  Fix this by recomputing
the deadline after *timeout could have changed.

With this change, ThreadPool still cannot participate in polling
but at least it does not suffer from extra latency.

Reported-by: Sergio Lopez <slp@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-id: 20190409122823.12416-1-pbonzini@redhat.com
Cc: Stefan Hajnoczi <stefanha@gmail.com>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: qemu-block@nongnu.org
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <1553692145-86728-1-git-send-email-pbonzini@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20190409122823.12416-1-pbonzini@redhat.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoblock/io.c: fix for the allocation failure
Andrey Shinkevich [Fri, 5 Apr 2019 14:24:04 +0000 (17:24 +0300)] 
block/io.c: fix for the allocation failure

On a file system used by the customer, fallocate() returns an error
if the block is not properly aligned. So, bdrv_co_pwrite_zeroes()
fails. We can handle that case the same way as it is done for the
unsupported cases, namely, call to bdrv_driver_pwritev() that writes
zeroes to an image for the unaligned chunk of the block.

Suggested-by: Denis V. Lunev <den@openvz.org>
Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Message-id: 1554474244-553661-1-git-send-email-andrey.shinkevich@virtuozzo.com
Message-Id: <1554474244-553661-1-git-send-email-andrey.shinkevich@virtuozzo.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoblock: Add coroutine_fn to bdrv_check_co_entry
Nikita Alekseev [Mon, 1 Apr 2019 09:30:51 +0000 (12:30 +0300)] 
block: Add coroutine_fn to bdrv_check_co_entry

bdrv_check_co_entry calls bdrv_co_check, which is a coroutine function.
Thus, it also needs to be marked as a coroutine.

Signed-off-by: Nikita Alekseev <n.alekseev2104@gmail.com>
Message-id: 20190401093051.16488-1-n.alekseev2104@gmail.com
Message-Id: <20190401093051.16488-1-n.alekseev2104@gmail.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoutil/readline: Add braces to fix checkpatch errors
Jules Irenge [Sat, 30 Mar 2019 11:21:42 +0000 (11:21 +0000)] 
util/readline: Add braces to fix checkpatch errors

Add braces to fix errors issued by checkpatch.pl tool
"ERROR: braces {} are necessary for all arms of this statement"
Within "util/readline.c" file
Message-Id: <20190330112142.14082-1-jbi.octave@gmail.com>

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoutil: readline: replace tab indent by four spaces to fix checkpatch errors
Jules Irenge [Mon, 1 Apr 2019 02:44:06 +0000 (03:44 +0100)] 
util: readline: replace tab indent by four spaces to fix checkpatch errors

Replace tab indent by four spaces to fix errors issued by checkpatch.pl tool
"ERROR: code indent should never use tabs" within "util/readline.c" file.

Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20190401024406.10819-3-jbi.octave@gmail.com
Message-Id: <20190401024406.10819-3-jbi.octave@gmail.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoutil/readline: add a space to fix errors by checkpatch tool
Jules Irenge [Mon, 1 Apr 2019 02:44:05 +0000 (03:44 +0100)] 
util/readline: add a space to fix errors by checkpatch tool

util/readline: add a space to fix errors reported by checkpatch.pl tool
"ERROR: space required before the open parenthesis"
"ERROR: space required after that ..."
within "util/redline.c" file

Signed-off-by: Jules Irenge <jbi.octave@gmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20190401024406.10819-2-jbi.octave@gmail.com
Message-Id: <20190401024406.10819-2-jbi.octave@gmail.com>
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
2 years agoMerge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2019-05-07' into staging
Peter Maydell [Thu, 9 May 2019 15:31:12 +0000 (16:31 +0100)] 
Merge remote-tracking branch 'remotes/maxreitz/tags/pull-block-2019-05-07' into staging

Block patches:
- Fixes to qcow2's implementation of qemu-img check
- Our SSH driver now supports bdrv_refresh_filename()
- Miscellaneous fixes

# gpg: Signature made Tue 07 May 2019 16:16:58 BST
# gpg:                using RSA key F407DB0061D5CF40
# gpg: Good signature from "Max Reitz <mreitz@redhat.com>" [full]
# Primary key fingerprint: 91BE B60A 30DB 3E88 57D1  1829 F407 DB00 61D5 CF40

* remotes/maxreitz/tags/pull-block-2019-05-07:
  iotests: Fix iotests 110 and 126
  commit: Use bdrv_append() in commit_start()
  block: Assert that drv->bdrv_child_perm is set in bdrv_child_perm()
  block/ssh: Implement .bdrv_dirname()
  block/ssh: Implement .bdrv_refresh_filename()
  qcow2: discard bitmap when removed
  qcow2-refcount: don't mask corruptions under internal errors
  qcow2-refcount: check_refcounts_l2: don't count fixed cluster as allocated
  qcow2-refcount: check_refcounts_l2: reduce ignored overlaps
  qcow2-refcount: avoid eating RAM
  qcow2-refcount: fix check_oflag_copied

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agolinux-user: avoid string truncation warnings in uname field copying
Daniel P. Berrangé [Wed, 1 May 2019 14:46:46 +0000 (15:46 +0100)] 
linux-user: avoid string truncation warnings in uname field copying

In file included from /usr/include/string.h:494,
                 from include/qemu/osdep.h:101,
                 from linux-user/uname.c:20:
In function ‘strncpy’,
    inlined from ‘sys_uname’ at linux-user/uname.c:94:3:
/usr/include/bits/string_fortified.h:106:10: warning: ‘__builtin_strncpy’ output may be truncated copying 64 bytes from a string of length 64 [-Wstringop-truncation]
  106 |   return __builtin___strncpy_chk (__dest, __src, __len, __bos (__dest));
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

We don't care where the NUL terminator in the original uname
field was. It suffices to copy the entire original field and
simply force a NUL terminator at the end of the new field.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20190501144646.4851-1-berrange@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agolinux-user/elfload: Fix GCC 9 build warnings
Alistair Francis [Tue, 30 Apr 2019 23:29:01 +0000 (23:29 +0000)] 
linux-user/elfload: Fix GCC 9 build warnings

Fix this warning when building with GCC9 on Fedora 30:
In function ‘strncpy’,
    inlined from ‘fill_psinfo’ at /home/alistair/qemu/linux-user/elfload.c:3208:12,
    inlined from ‘fill_note_info’ at /home/alistair/qemu/linux-user/elfload.c:3390:5,
    inlined from ‘elf_core_dump’ at /home/alistair/qemu/linux-user/elfload.c:3539:9:
/usr/include/bits/string_fortified.h:106:10: error: ‘__builtin_strncpy’ specified bound 16 equals destination size [-Werror=stringop-truncation]
  106 |   return __builtin___strncpy_chk (__dest, __src, __len, __bos (__dest));
      |          ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Signed-off-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Laurent Vivier <laurent@vivier.eu>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <c4d2b1de9efadcf1c900b91361af9302823a72a9.1556666645.git.alistair.francis@wdc.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agolinux-user: Add missing IPV6 sockopts
Helge Deller [Sun, 10 Mar 2019 11:12:16 +0000 (12:12 +0100)] 
linux-user: Add missing IPV6 sockopts

When running ssh over IPv6 with linux-user I faced this warning:
 Unsupported setsockopt level=41 optname=67
 setsockopt IPV6_TCLASS 32: Protocol not available:

This patch adds code to the linux-user emulatation for setting and
retrieving of a few missing IPV6 options, including IPV6_TCLASS.

Signed-off-by: Helge Deller <deller@gmx.de>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
2 years agoMerge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2019-05-09' into...
Peter Maydell [Thu, 9 May 2019 14:43:41 +0000 (15:43 +0100)] 
Merge remote-tracking branch 'remotes/huth-gitlab/tags/pull-request-2019-05-09' into staging

- Fix "make check" problem that occurred with LANG=C and Python 3.5 / 3.6
- Get rid of some more dependencies on the global_qtest variable in the qtests
- Some other small test clean-ups
- Some copyright statement clarifications
- Mark TARGET_FMT_lu as poisoned

# gpg: Signature made Thu 09 May 2019 08:45:47 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-2019-05-09:
  include/exec/poison: Mark TARGET_FMT_lu as poisoned, too
  target/sh4: Fix LGPL information in the file headers
  target/openrisc: Fix LGPL information in the file headers
  hw/i2c/smbus_ich9: Fix the confusing contributions-after-2012 statement
  tests: qpci_unplug_acpi_device_test() should not rely on global_qtest
  tests/drive_del-test: Use qtest_init() instead of qtest_start()
  tests/Makefile: Remove unused test-obj-y variable
  tests/tpm-tests: Use g_test_skip() to mark skipped tests
  tests/ide-test: Make test independent of global_qtest
  tests/test-hmp: Use qtest_init() instead of qtest_start()
  tests/qmp-cmd-test: Use qtest_init() instead of qtest_start()
  tests/megasas: Make test independent of global_qtest
  tests/tco: Make test independent of global_qtest
  tests: Force Python I/O encoding for check-qapi-schema

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-pull-request' into...
Peter Maydell [Thu, 9 May 2019 12:36:10 +0000 (13:36 +0100)] 
Merge remote-tracking branch 'remotes/vivier2/tags/trivial-branch-pull-request' into staging

Pull request trivial branch 2019-05-03

# gpg: Signature made Fri 03 May 2019 12:26:34 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:
  sockets: avoid string truncation warnings when copying UNIX path
  hw/sparc/leon3: Allow load of uImage firmwares
  Makefile: Let the 'clean' rule remove qemu-ga.exe on Windows hosts
  net: Print output of "-net nic, model=help" to stdout instead of stderr
  Header cleanups
  Update configure
  configure: fix pam test warning
  qom: use object_new_with_type in object_new_with_propv
  doc: fix the configuration path
  CODING_STYLE: indent example code as all others
  CODING_STYLE: specify the indent rule for multiline code
  hw/net/pcnet: Use qemu_log_mask(GUEST_ERROR) instead of printf

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging
Peter Maydell [Thu, 9 May 2019 08:58:41 +0000 (09:58 +0100)] 
Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging

Update slirp submodule

Samuel Thibault (1):
  Update upstream slirp
  Adds gitignore, README file, and fixes ident protocol parsing.

# gpg: Signature made Thu 09 May 2019 09:00:30 BST
# gpg:                using RSA key E61DBB15D4172BDEC97E92D9DB550E89F0FA54F3
# gpg: Good signature from "Samuel Thibault <samuel.thibault@aquilenet.fr>" [unknown]
# gpg:                 aka "Samuel Thibault <sthibault@debian.org>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@gnu.org>" [unknown]
# gpg:                 aka "Samuel Thibault <samuel.thibault@inria.fr>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@labri.fr>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@ens-lyon.org>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@u-bordeaux.fr>" [unknown]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 900C B024 B679 31D4 0F82  304B D017 8C76 7D06 9EE6
#      Subkey fingerprint: E61D BB15 D417 2BDE C97E  92D9 DB55 0E89 F0FA 54F3

* remotes/thibault/tags/samuel-thibault:
  Update upstream slirp

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoUpdate upstream slirp
Samuel Thibault [Thu, 9 May 2019 07:57:23 +0000 (09:57 +0200)] 
Update upstream slirp

Adds gitignore, README file, and fixes ident protocol parsing.

Signed-off-by: Samuel Thibault <samuel.thibault@ens-lyon.org>
2 years agoinclude/exec/poison: Mark TARGET_FMT_lu as poisoned, too
Thomas Huth [Mon, 6 May 2019 04:05:42 +0000 (06:05 +0200)] 
include/exec/poison: Mark TARGET_FMT_lu as poisoned, too

We already poison TARGET_FMT_lx and TARGET_FMT_ld, but apparently
forgot to poison TARGET_FMT_lu, too. Do it now.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20190508150608.3311-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotarget/sh4: Fix LGPL information in the file headers
Thomas Huth [Wed, 13 Feb 2019 13:52:50 +0000 (14:52 +0100)] 
target/sh4: Fix LGPL information in the file headers

It's either "GNU *Library* General Public License version 2" or "GNU
Lesser General Public License version *2.1*", but there was no "version
2.0" of the "Lesser" license. So assume that version 2.1 is meant here.

Message-Id: <1550073530-4138-1-git-send-email-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotarget/openrisc: Fix LGPL information in the file headers
Thomas Huth [Wed, 13 Feb 2019 13:46:50 +0000 (14:46 +0100)] 
target/openrisc: Fix LGPL information in the file headers

It's either "GNU *Library* General Public License version 2" or "GNU
Lesser General Public License version *2.1*", but there was no "version
2.0" of the "Lesser" license. So assume that version 2.1 is meant here.

Acked-by: Stafford Horne <shorne@gmail.com>
Message-Id: <1550073577-4248-1-git-send-email-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agohw/i2c/smbus_ich9: Fix the confusing contributions-after-2012 statement
Thomas Huth [Wed, 6 Feb 2019 16:33:38 +0000 (17:33 +0100)] 
hw/i2c/smbus_ich9: Fix the confusing contributions-after-2012 statement

The license information in this file is rather confusing. The text
declares LGPL first, but then says that contributions after Jan 2012
are licensed under the GPL instead. How should the average user who
just downloaded the release tarball know which part is now GPL and
which is LGPL? Also, as far as I can see, the file has been added to
QEMU *after* January in 2012, so the whole file should be GPL by
default instead.

Furthermore, looking at the text of the LGPL (see COPYING.LIB in the
top directory), the license clearly states in section "3." that one
should rather replace the license information in such a case instead.
Thus let's clean up the confusing statements and use the proper GPL
text only.

Message-Id: <1549471435-21887-1-git-send-email-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests: qpci_unplug_acpi_device_test() should not rely on global_qtest
Thomas Huth [Tue, 9 Apr 2019 12:25:57 +0000 (14:25 +0200)] 
tests: qpci_unplug_acpi_device_test() should not rely on global_qtest

libqos functions should not use functions that require global_qtest to
be set, since such library functions could also be used by tests that
deal with multiple test states. Add a parameter to this function to
explicitly specify the test state.

Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20190508143209.24350-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/drive_del-test: Use qtest_init() instead of qtest_start()
Thomas Huth [Tue, 9 Apr 2019 10:38:41 +0000 (12:38 +0200)] 
tests/drive_del-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.

Reviewed-by: Eric Blake <eblake@redhat.com>
Message-Id: <20190508142153.21555-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/Makefile: Remove unused test-obj-y variable
Thomas Huth [Wed, 8 May 2019 07:55:27 +0000 (09:55 +0200)] 
tests/Makefile: Remove unused test-obj-y variable

I recently noticed that test-obj-y contains a file called
tests/check-block-qtest.o which simply does not belong to any .c
file and thus wondered why this is not causing any trouble. It is
only used to add -Itests to the command line (which refers to the
build directory). However, it is not needed because "-iquote $(@D)"
already sets this up in rules.mak. Thus we can simply remove this
variable.

Reviewed-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190508075527.32164-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agoMerge remote-tracking branch 'remotes/ericb/tags/pull-nbd-2019-05-07' into staging
Peter Maydell [Wed, 8 May 2019 15:10:24 +0000 (16:10 +0100)] 
Merge remote-tracking branch 'remotes/ericb/tags/pull-nbd-2019-05-07' into staging

nbd patches for 2019-05-07

- iotest improvements

# gpg: Signature made Tue 07 May 2019 16:00:18 BST
# gpg:                using RSA key A7A16B4A2527436A
# gpg: Good signature from "Eric Blake <eblake@redhat.com>" [full]
# gpg:                 aka "Eric Blake (Free Software Programmer) <ebb9@byu.net>" [full]
# gpg:                 aka "[jpeg image of size 6874]" [full]
# Primary key fingerprint: 71C2 CC22 B1C4 6029 27D2  F3AA A7A1 6B4A 2527 436A

* remotes/ericb/tags/pull-nbd-2019-05-07:
  iotests: Make 182 do without device_add
  iotests: Tweak 221 sizing for different hole granularities
  tests/qemu-iotests: Fix more reference output files due to recent qemu-io change
  iotests: Fix 233 for ports other than 10809
  qemu-nbd: Look up flag names in array

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agotests/tpm-tests: Use g_test_skip() to mark skipped tests
Thomas Huth [Wed, 24 Apr 2019 09:43:13 +0000 (11:43 +0200)] 
tests/tpm-tests: Use g_test_skip() to mark skipped tests

Since we do not use gtester anymore (which had a bug here),
we can now use g_test_skip() to mark skipped tests.

Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Stefan Berger <stefanb@linux.ibm.com>
Message-Id: <20190424094557.28404-1-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/ide-test: Make test independent of global_qtest
Thomas Huth [Tue, 9 Apr 2019 08:52:45 +0000 (10:52 +0200)] 
tests/ide-test: Make test independent of global_qtest

Pass around the QTestState, so we do not need the problematic global_qtest
variable (which causes trouble for tests that have multiple test states)
here anymore.

Message-Id: <20190409085245.31548-6-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/test-hmp: Use qtest_init() instead of qtest_start()
Thomas Huth [Tue, 9 Apr 2019 08:52:44 +0000 (10:52 +0200)] 
tests/test-hmp: 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.

Reviewed-by: Dr. David Alan Gilbert <dgilbert@redhat.com>
Message-Id: <20190409085245.31548-5-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/qmp-cmd-test: Use qtest_init() instead of qtest_start()
Thomas Huth [Tue, 9 Apr 2019 08:52:43 +0000 (10:52 +0200)] 
tests/qmp-cmd-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
(since global_qtest can not be used in tests that have to track
multiple QEMU states, like migration tests). Use qtest_init() and
qtest_quit() instead.

Message-Id: <20190409085245.31548-4-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/megasas: Make test independent of global_qtest
Thomas Huth [Tue, 9 Apr 2019 08:52:42 +0000 (10:52 +0200)] 
tests/megasas: Make test independent of global_qtest

The test uses memwrite() and thus relies on global_qtest. Let's replace it
with qtest_memwrite(), so that we are independent from global_qtest here.

Message-Id: <20190409085245.31548-3-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests/tco: Make test independent of global_qtest
Thomas Huth [Tue, 9 Apr 2019 08:52:41 +0000 (10:52 +0200)] 
tests/tco: Make test independent of global_qtest

Pass around the QTestState in the TestData, so we do not need the
global_qtest variable here anymore.

Message-Id: <20190409085245.31548-2-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agotests: Force Python I/O encoding for check-qapi-schema
Eduardo Habkost [Mon, 6 May 2019 21:38:17 +0000 (18:38 -0300)] 
tests: Force Python I/O encoding for check-qapi-schema

test-qapi.py doesn't force a specific encoding for stderr or
stdout, but the reference files used by check-qapi-schema are in
UTF-8.  This breaks check-qapi-schema under certain circumstances
(e.g. if using the C locale and Python < 3.7).

We need to make sure test-qapi.py always generate UTF-8 output
somehow.  On Python 3.7+ we can do it using
`sys.stdout.reconfigure(...)`, but we need a solution that works
with older Python versions.

Instead of trying a hack like reopening sys.stdout and
sys.stderr, we can just tell Python to use UTF-8 for I/O encoding
when running test-qapi.py.  Do it by setting PYTHONIOENCODING.

Reported-by: Thomas Huth <thuth@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Message-Id: <20190506213817.14344-1-ehabkost@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Tested-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
2 years agoMerge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20190507' into...
Peter Maydell [Tue, 7 May 2019 23:06:43 +0000 (00:06 +0100)] 
Merge remote-tracking branch 'remotes/pmaydell/tags/pull-target-arm-20190507' into staging

target-arm queue:
 * Stop using variable length array in dc_zva
 * Implement M-profile XPSR GE bits
 * Don't enable ARMV7M_EXCP_DEBUG from reset
 * armv7m_nvic: NS BFAR and BFSR are RAZ/WI if BFHFNMINS == 0
 * armv7m_nvic: Check subpriority in nvic_recompute_state_secure()
 * fix various minor issues to allow building for Windows-on-ARM64
 * aspeed: Set SDRAM size
 * Allow system registers for KVM guests to be changed by QEMU code
 * raspi: Diagnose requests for too much RAM
 * virt: Support firmware configuration with -blockdev

# gpg: Signature made Tue 07 May 2019 12:59:30 BST
# gpg:                using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg:                issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [ultimate]
# gpg:                 aka "Peter Maydell <pmaydell@gmail.com>" [ultimate]
# gpg:                 aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [ultimate]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83  15CF 3C25 25ED 1436 0CDE

* remotes/pmaydell/tags/pull-target-arm-20190507:
  target/arm: Stop using variable length array in dc_zva
  target/arm: Implement XPSR GE bits
  hw/intc/armv7m_nvic: Don't enable ARMV7M_EXCP_DEBUG from reset
  hw/intc/armv7m_nvic: NS BFAR and BFSR are RAZ/WI if BFHFNMINS == 0
  hw/arm/armv7m_nvic: Check subpriority in nvic_recompute_state_secure()
  osdep: Fix mingw compilation regarding stdio formats
  util/cacheinfo: Use uint64_t on LLP64 model to satisfy Windows ARM64
  qga: Fix mingw compilation warnings on enum conversion
  QEMU_PACKED: Remove gcc_struct attribute in Windows non x86 targets
  arm: aspeed: Set SDRAM size
  arm: Allow system registers for KVM guests to be changed by QEMU code
  hw/arm/raspi: Diagnose requests for too much RAM
  hw/arm/virt: Support firmware configuration with -blockdev
  pflash_cfi01: New pflash_cfi01_legacy_drive()
  pc: Rearrange pc_system_firmware_init()'s legacy -drive loop

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/kraxel/tags/vga-20190507-pull-request' into...
Peter Maydell [Tue, 7 May 2019 20:39:28 +0000 (21:39 +0100)] 
Merge remote-tracking branch 'remotes/kraxel/tags/vga-20190507-pull-request' into staging

vga: bugfixes for qxl, cirrus, ati.
vga: add "-vga help" support.
vga: move i2c-ddc to display.

# gpg: Signature made Tue 07 May 2019 09:19:32 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/vga-20190507-pull-request:
  i2c-ddc: move it to hw/display
  ati-vga: Fix check for blt outside vram
  qxl: avoid unaligned pointer reads/writes
  vl: add -vga help support
  vl: constify VGAInterfaceInfo
  hw/display/cirrus_vga: Remove unused include
  hw/display/cirrus_vga: Update the documentation URL
  qxl: check release info object

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/rth/tags/pull-dt-20190506' into staging
Peter Maydell [Tue, 7 May 2019 16:16:11 +0000 (17:16 +0100)] 
Merge remote-tracking branch 'remotes/rth/tags/pull-dt-20190506' into staging

Add support for variable-length ISAs

# gpg: Signature made Mon 06 May 2019 19:26:21 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-dt-20190506:
  decodetree: Add DisasContext argument to !function expanders
  decodetree: Expand a decode_load function
  decodetree: Initial support for variable-length ISAs

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoiotests: Fix iotests 110 and 126
Max Reitz [Mon, 15 Apr 2019 15:41:29 +0000 (17:41 +0200)] 
iotests: Fix iotests 110 and 126

A recent patch results in qemu-img reporting the backing file format of
vmdk images as vmdk.  This broke iotests 110 and 126.

Fixes: 7502be838e2fb62cc00f9e55f632e0b88ead5e6e
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190415154129.31021-1-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agocommit: Use bdrv_append() in commit_start()
Alberto Garcia [Wed, 3 Apr 2019 14:37:48 +0000 (17:37 +0300)] 
commit: Use bdrv_append() in commit_start()

This function combines bdrv_set_backing_hd() and bdrv_replace_node()
so we can use it to simplify the code a bit in commit_start().

Signed-off-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190403143748.9790-1-berto@igalia.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoblock: Assert that drv->bdrv_child_perm is set in bdrv_child_perm()
Alberto Garcia [Thu, 4 Apr 2019 11:29:53 +0000 (14:29 +0300)] 
block: Assert that drv->bdrv_child_perm is set in bdrv_child_perm()

There is no need to check for this because all block drivers that have
children implement bdrv_child_perm and all callers already ensure that
bs->drv is set.

Furthermore, if this check would fail then the callers would end up
with uninitialized values for nperm and nshared.

This patch replaces the check with an assertion.

Signed-off-by: Alberto Garcia <berto@igalia.com>
Message-id: 20190404112953.4058-1-berto@igalia.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoblock/ssh: Implement .bdrv_dirname()
Max Reitz [Mon, 25 Feb 2019 19:08:28 +0000 (20:08 +0100)] 
block/ssh: Implement .bdrv_dirname()

ssh_bdrv_dirname() is basically the generic bdrv_dirname(), except it
takes care not to silently chop off any query string (i.e.,
host_key_check).

Signed-off-by: Max Reitz <mreitz@redhat.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Message-id: 20190225190828.17726-3-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoblock/ssh: Implement .bdrv_refresh_filename()
Max Reitz [Mon, 25 Feb 2019 19:08:27 +0000 (20:08 +0100)] 
block/ssh: Implement .bdrv_refresh_filename()

This requires some changes to keep iotests 104 and 207 working.

qemu-img info in 104 will now return a filename including the user name
and the port, which need to be filtered by adjusting REMOTE_TEST_DIR in
common.rc.  This additional information has to be marked optional,
however (which is simple as REMOTE_TEST_DIR is a regex), because
otherwise 197 and 215 would fail: They use it (indirectly) to filter
qemu-img create output which contains a backing filename they have
passed to it -- which probably does not contain a user name or port
number.

The problem in 207 is a nice one to have: qemu-img info used to return
json:{} filenames, but with this patch it returns nice plain ones.  We
now need to adjust the filtering to hide the user name (and port number
while we are at it).  The simplest way to do this is to include both in
iotests.remote_filename() so that bdrv_refresh_filename() will not
change it, and then iotests.img_info_log() will filter it correctly
automatically.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Tested-by: Richard W.M. Jones <rjones@redhat.com>
Message-id: 20190225190828.17726-2-mreitz@redhat.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoqcow2: discard bitmap when removed
Andrey Shinkevich [Thu, 28 Feb 2019 09:26:59 +0000 (12:26 +0300)] 
qcow2: discard bitmap when removed

Bitmap data may take a lot of disk space, so it's better to discard it
always.

Signed-off-by: Andrey Shinkevich <andrey.shinkevich@virtuozzo.com>
Message-id: 1551346019-293202-1-git-send-email-andrey.shinkevich@virtuozzo.com
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
[mreitz: Use the commit message proposed by Vladimir]
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoqcow2-refcount: don't mask corruptions under internal errors
Vladimir Sementsov-Ogievskiy [Wed, 27 Feb 2019 13:14:33 +0000 (16:14 +0300)] 
qcow2-refcount: don't mask corruptions under internal errors

No reasons for not reporting found corruptions as corruptions in case
of some internal errors, especially in case of just failed to fix l2
entry (and in this case, missed corruptions may influence comparing
logic, when we calculate difference between corruptions fields of two
results)

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Message-id: 20190227131433.197063-6-vsementsov@virtuozzo.com
Reviewed-by: Max Reitz <mreitz@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoqcow2-refcount: check_refcounts_l2: don't count fixed cluster as allocated
Vladimir Sementsov-Ogievskiy [Wed, 27 Feb 2019 13:14:32 +0000 (16:14 +0300)] 
qcow2-refcount: check_refcounts_l2: don't count fixed cluster as allocated

Do not count a cluster which is fixed to be ZERO as allocated.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-5-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoqcow2-refcount: check_refcounts_l2: reduce ignored overlaps
Vladimir Sementsov-Ogievskiy [Wed, 27 Feb 2019 13:14:31 +0000 (16:14 +0300)] 
qcow2-refcount: check_refcounts_l2: reduce ignored overlaps

Reduce number of structures ignored in overlap check: when checking
active table ignore active tables, when checking inactive table ignore
inactive ones.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-4-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoqcow2-refcount: avoid eating RAM
Vladimir Sementsov-Ogievskiy [Wed, 27 Feb 2019 13:14:30 +0000 (16:14 +0300)] 
qcow2-refcount: avoid eating RAM

qcow2_inc_refcounts_imrt() (through realloc_refcount_array()) can eat
an unpredictable amount of memory on corrupted table entries, which are
referencing regions far beyond the end of file.

Prevent this, by skipping such regions from further processing.

Interesting that iotest 138 checks exactly the behavior which we fix
here. So, change the test appropriately.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-3-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoqcow2-refcount: fix check_oflag_copied
Vladimir Sementsov-Ogievskiy [Wed, 27 Feb 2019 13:14:29 +0000 (16:14 +0300)] 
qcow2-refcount: fix check_oflag_copied

Increase corruptions_fixed only after successful fix.

Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Reviewed-by: Max Reitz <mreitz@redhat.com>
Message-id: 20190227131433.197063-2-vsementsov@virtuozzo.com
Signed-off-by: Max Reitz <mreitz@redhat.com>
2 years agoiotests: Make 182 do without device_add
Max Reitz [Wed, 17 Apr 2019 15:30:05 +0000 (17:30 +0200)] 
iotests: Make 182 do without device_add

182 fails if qemu has no support for hotplugging of a virtio-blk device.
Using an NBD server instead works just as well for the test, even on
qemus without hotplugging support.

Fixes: 6d0a4a0fb5c8f10c8eb68b52cfda0082b00ae963
Reported-by: Danilo C. L. de Paula <ddepaula@redhat.com>
Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20190417153005.30096-1-mreitz@redhat.com>
Tested-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
Acked-by: Alberto Garcia <berto@igalia.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2 years agoiotests: Tweak 221 sizing for different hole granularities
Eric Blake [Mon, 6 May 2019 17:21:11 +0000 (12:21 -0500)] 
iotests: Tweak 221 sizing for different hole granularities

For some particular configurations of ext4, sizing an image to 84
sectors + 1 byte causes test failures when the size of the hole is
rounded to a 4k alignment. Let's instead size things to 128 sectors +
1 byte, as the 64k boundary is more likely to work with various hole
granularities.

Reported-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
Message-Id: <20190506172111.31594-1-eblake@redhat.com>
Tested-by: Thomas Huth <thuth@redhat.com>
2 years agotests/qemu-iotests: Fix more reference output files due to recent qemu-io change
Thomas Huth [Wed, 1 May 2019 13:41:27 +0000 (15:41 +0200)] 
tests/qemu-iotests: Fix more reference output files due to recent qemu-io change

The output of qemu-io changed recently - most tests have been fixed in
commit 36b9986b08787019ef42 ("tests/qemu-iotests: Fix output of qemu-io
related tests") already, but a qcow1, vmdk, and nbd test were still missing.

Fixes: 99e98d7c9fc1a1639fad ("qemu-io: Use error_[gs]et_progname()")
Signed-off-by: Thomas Huth <thuth@redhat.com>
Message-Id: <20190501134127.21104-1-thuth@redhat.com>
[eblake: squash in NBD 083 fixes]
Reviewed-by: Stefano Garzarella <sgarzare@redhat.com>
Tested-by: Stefano Garzarella <sgarzare@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2 years agoiotests: Fix 233 for ports other than 10809
Max Reitz [Mon, 6 May 2019 16:05:29 +0000 (18:05 +0200)] 
iotests: Fix 233 for ports other than 10809

233 generally filters the port, but in two cases does not.  If some
other concurrently running application has already taken port 10809,
this will result in an output mismatch.  Fix this by applying the
filter in these two cases, too.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20190506160529.6955-1-mreitz@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2 years agoqemu-nbd: Look up flag names in array
Max Reitz [Fri, 5 Apr 2019 19:16:35 +0000 (21:16 +0200)] 
qemu-nbd: Look up flag names in array

The existing code to convert flag bits into strings looks a bit strange
now, and if we ever add more flags, it will look even stranger.  Prevent
that from happening by making it look up the flag names in an array.

Signed-off-by: Max Reitz <mreitz@redhat.com>
Message-Id: <20190405191635.25740-1-mreitz@redhat.com>
Signed-off-by: Eric Blake <eblake@redhat.com>
2 years agotarget/arm: Stop using variable length array in dc_zva
Peter Maydell [Tue, 7 May 2019 11:55:04 +0000 (12:55 +0100)] 
target/arm: Stop using variable length array in dc_zva

Currently the dc_zva helper function uses a variable length
array. In fact we know (as the comment above remarks) that
the length of this array is bounded because the architecture
limits the block size and QEMU limits the target page size.
Use a fixed array size and assert that we don't run off it.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190503120448.13385-1-peter.maydell@linaro.org

2 years agotarget/arm: Implement XPSR GE bits
Peter Maydell [Tue, 7 May 2019 11:55:04 +0000 (12:55 +0100)] 
target/arm: Implement XPSR GE bits

In the M-profile architecture, if the CPU implements the DSP extension
then the XPSR has GE bits, in the same way as the A-profile CPSR. When
we added DSP extension support we forgot to add support for reading
and writing the GE bits, which are stored in env->GE. We did put in
the code to add XPSR_GE to the mask of bits to update in the v7m_msr
helper, but forgot it in v7m_mrs. We also must not allow the XPSR we
pull off the stack on exception return to set the nonexistent GE bits.
Correct these errors:
 * read and write env->GE in xpsr_read() and xpsr_write()
 * only set GE bits on exception return if DSP present
 * read GE bits for MRS if DSP present

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20190430131439.25251-5-peter.maydell@linaro.org

2 years agohw/intc/armv7m_nvic: Don't enable ARMV7M_EXCP_DEBUG from reset
Peter Maydell [Tue, 7 May 2019 11:55:03 +0000 (12:55 +0100)] 
hw/intc/armv7m_nvic: Don't enable ARMV7M_EXCP_DEBUG from reset

The M-profile architecture specifies that the DebugMonitor exception
should be initially disabled, not enabled. It should be controlled
by the DEMCR register's MON_EN bit, but we don't implement that
register yet (like most of the debug architecture for M-profile).

Note that BKPT instructions will still work, because they
will be escalated to HardFault.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20190430131439.25251-4-peter.maydell@linaro.org

2 years agohw/intc/armv7m_nvic: NS BFAR and BFSR are RAZ/WI if BFHFNMINS == 0
Peter Maydell [Tue, 7 May 2019 11:55:03 +0000 (12:55 +0100)] 
hw/intc/armv7m_nvic: NS BFAR and BFSR are RAZ/WI if BFHFNMINS == 0

The non-secure versions of the BFAR and BFSR registers are
supposed to be RAZ/WI if AICR.BFHFNMINS == 0; we were
incorrectly allowing NS code to access the real values.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20190430131439.25251-3-peter.maydell@linaro.org

2 years agohw/arm/armv7m_nvic: Check subpriority in nvic_recompute_state_secure()
Peter Maydell [Tue, 7 May 2019 11:55:03 +0000 (12:55 +0100)] 
hw/arm/armv7m_nvic: Check subpriority in nvic_recompute_state_secure()

Rule R_CQRV says that if two pending interrupts have the same
group priority then ties are broken by looking at the subpriority.
We had a comment describing this but had forgotten to actually
implement the subpriority comparison. Correct the omission.

(The further tie break rules of "lowest exception number" and
"secure before non-secure" are handled implicitly by the order
in which we iterate through the exceptions in the loops.)

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20190430131439.25251-2-peter.maydell@linaro.org

2 years agoosdep: Fix mingw compilation regarding stdio formats
Cao Jiaxi [Tue, 7 May 2019 11:55:03 +0000 (12:55 +0100)] 
osdep: Fix mingw compilation regarding stdio formats

I encountered the following compilation error on mingw:

/mnt/d/qemu/include/qemu/osdep.h:97:9: error: '__USE_MINGW_ANSI_STDIO' macro redefined [-Werror,-Wmacro-redefined]
 #define __USE_MINGW_ANSI_STDIO 1
        ^
/mnt/d/llvm-mingw/aarch64-w64-mingw32/include/_mingw.h:433:9: note: previous definition is here
 #define __USE_MINGW_ANSI_STDIO 0      /* was not defined so it should be 0 */

It turns out that __USE_MINGW_ANSI_STDIO must be set before any
system headers are included, not just before stdio.h.

Signed-off-by: Cao Jiaxi <driver1998@foxmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Stefan Weil <sw@weilnetz.de>
Message-id: 20190503003719.10233-1-driver1998@foxmail.com
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoutil/cacheinfo: Use uint64_t on LLP64 model to satisfy Windows ARM64
Cao Jiaxi [Tue, 7 May 2019 11:55:03 +0000 (12:55 +0100)] 
util/cacheinfo: Use uint64_t on LLP64 model to satisfy Windows ARM64

Windows ARM64 uses LLP64 model, which breaks current assumptions.

Signed-off-by: Cao Jiaxi <driver1998@foxmail.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20190503003707.10185-1-driver1998@foxmail.com
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoqga: Fix mingw compilation warnings on enum conversion
Cao Jiaxi [Tue, 7 May 2019 11:55:03 +0000 (12:55 +0100)] 
qga: Fix mingw compilation warnings on enum conversion

The win2qemu[] is supposed to be the conversion table to convert between
STORAGE_BUS_TYPE in Windows SDK and GuestDiskBusType in qga.

But it was incorrectly written that it forces to set a GuestDiskBusType
value to STORAGE_BUS_TYPE, which generates an enum conversion warning in clang.

Suggested-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Cao Jiaxi <driver1998@foxmail.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20190503003650.10137-1-driver1998@foxmail.com
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoQEMU_PACKED: Remove gcc_struct attribute in Windows non x86 targets
Cao Jiaxi [Tue, 7 May 2019 11:55:02 +0000 (12:55 +0100)] 
QEMU_PACKED: Remove gcc_struct attribute in Windows non x86 targets

gcc_struct is for x86 only, and it generates an warning on ARM64 Clang/MinGW targets.

Signed-off-by: Cao Jiaxi <driver1998@foxmail.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20190503003618.10089-1-driver1998@foxmail.com
[PMM: dropped the slirp change as slirp is now a submodule]
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoarm: aspeed: Set SDRAM size
Joel Stanley [Tue, 7 May 2019 11:55:02 +0000 (12:55 +0100)] 
arm: aspeed: Set SDRAM size

We currently use Qemu's default of 128MB. As we know how much ram each
machine ships with, make it easier on users by setting a default.

It can still be overridden with -m on the command line.

Signed-off-by: Joel Stanley <joel@jms.id.au>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20190503022958.1394-1-joel@jms.id.au
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoarm: Allow system registers for KVM guests to be changed by QEMU code
Peter Maydell [Tue, 7 May 2019 11:55:02 +0000 (12:55 +0100)] 
arm: Allow system registers for KVM guests to be changed by QEMU code

At the moment the Arm implementations of kvm_arch_{get,put}_registers()
don't support having QEMU change the values of system registers
(aka coprocessor registers for AArch32). This is because although
kvm_arch_get_registers() calls write_list_to_cpustate() to
update the CPU state struct fields (so QEMU code can read the
values in the usual way), kvm_arch_put_registers() does not
call write_cpustate_to_list(), meaning that any changes to
the CPU state struct fields will not be passed back to KVM.

The rationale for this design is documented in a comment in the
AArch32 kvm_arch_put_registers() -- writing the values in the
cpregs list into the CPU state struct is "lossy" because the
write of a register might not succeed, and so if we blindly
copy the CPU state values back again we will incorrectly
change register values for the guest. The assumption was that
no QEMU code would need to write to the registers.

However, when we implemented debug support for KVM guests, we
broke that assumption: the code to handle "set the guest up
to take a breakpoint exception" does so by updating various
guest registers including ESR_EL1.

Support this by making kvm_arch_put_registers() synchronize
CPU state back into the list. We sync only those registers
where the initial write succeeds, which should be sufficient.

This commit is the same as commit 823e1b3818f9b10b824ddc which we
had to revert in commit 942f99c825fc94c8b1a4, except that the bug
which was preventing EDK2 guest firmware running has been fixed:
kvm_arm_reset_vcpu() now calls write_list_to_cpustate().

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Eric Auger <eric.auger@redhat.com>
2 years agohw/arm/raspi: Diagnose requests for too much RAM
Peter Maydell [Tue, 7 May 2019 11:55:02 +0000 (12:55 +0100)] 
hw/arm/raspi: Diagnose requests for too much RAM

The Raspberry Pi boards have a physical memory map which does
not allow for more than 1GB of RAM. Currently if the user tries
to ask for more then we fail in a confusing way:

$ qemu-system-aarch64 --machine raspi3 -m 8G
Unexpected error in visit_type_uintN() at qapi/qapi-visit-core.c:164:
qemu-system-aarch64: Parameter 'vcram-base' expects uint32_t
Aborted (core dumped)

Catch this earlier and diagnose it with a more friendly message:
$ qemu-system-aarch64 --machine raspi3 -m 8G
qemu-system-aarch64: Requested ram size is too large for this machine: maximum is 1GB

Fixes: https://bugs.launchpad.net/qemu/+bug/1794187
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Wainer dos Santos Moschetta <wainersm@redhat.com>
2 years agohw/arm/virt: Support firmware configuration with -blockdev
Markus Armbruster [Tue, 7 May 2019 11:55:02 +0000 (12:55 +0100)] 
hw/arm/virt: Support firmware configuration with -blockdev

The ARM virt machines put firmware in flash memory.  To configure it,
you use -drive if=pflash,unit=0,... and optionally -drive
if=pflash,unit=1,...

Why two -drive?  This permits setting up one part of the flash memory
read-only, and the other part read/write.  It also makes upgrading
firmware on the host easier.  Below the hood, we get two separate
flash devices, because we were too lazy to improve our flash device
models to support sector protection.

The problem at hand is to do the same with -blockdev somehow, as one
more step towards deprecating -drive.

We recently solved this problem for x86 PC machines, in commit
ebc29e1beab.  See the commit message for design rationale.

This commit solves it for ARM virt basically the same way: new machine
properties pflash0, pflash1 forward to the onboard flash devices'
properties.  Requires creating the onboard devices in the
.instance_init() method virt_instance_init().  The existing code to
pick up drives defined with -drive if=pflash is replaced by code to
desugar into the machine properties.

There are a few behavioral differences, though:

* The flash devices are always present (x86: only present if
  configured)

* Flash base addresses and sizes are fixed (x86: sizes depend on
  images, mapped back to back below a fixed address)

* -bios configures contents of first pflash (x86: -bios configures ROM
   contents)

* -bios is rejected when first pflash is also configured with -machine
   pflash0=... (x86: bios is silently ignored then)

* -machine pflash1=... does not require -machine pflash0=... (x86: it
   does).

The actual code is a bit simpler than for x86 mostly due to the first
two differences.

Before the patch, all the action is in create_flash(), called from the
machine's .init() method machvirt_init():

    main()
        machine_run_board_init()
            machvirt_init()
                create_flash()
                    create_one_flash() for flash[0]
                        create
                        configure
                            includes obeying -drive if=pflash,unit=0
                        realize
                        map
                        fall back to -bios
                    create_one_flash() for flash[1]
                        create
                        configure
                            includes obeying -drive if=pflash,unit=1
                        realize
                        map
                    update FDT

To make the machine properties work, we need to move device creation
to its .instance_init() method virt_instance_init().

Another complication is machvirt_init()'s computation of
@firmware_loaded: it predicts what create_flash() will do.  Instead of
predicting what create_flash()'s replacement virt_firmware_init() will
do, I decided to have virt_firmware_init() return what it did.
Requires calling it a bit earlier.

Resulting call tree:

    main()
        current_machine = object_new()
            ...
                virt_instance_init()
                    virt_flash_create()
                        virt_flash_create1() for flash[0]
                            create
                            configure: set defaults
                            become child of machine [NEW]
                            add machine prop pflash0 as alias for drive [NEW]
                        virt_flash_create1() for flash[1]
                            create
                            configure: set defaults
                            become child of machine [NEW]
                            add machine prop pflash1 as alias for drive [NEW]
        for all machine props from the command line: machine_set_property()
            ...
                property_set_alias() for machine props pflash0, pflash1
                    ...
                        set_drive() for cfi.pflash01 prop drive
                            this is how -machine pflash0=... etc set
        machine_run_board_init(current_machine);
            virt_firmware_init()
                pflash_cfi01_legacy_drive()
                    legacy -drive if=pflash,unit=0 and =1 [NEW]
                virt_flash_map()
                    virt_flash_map1() for flash[0]
                        configure: num-blocks
                        realize
                        map
                    virt_flash_map1() for flash[1]
                        configure: num-blocks
                        realize
                        map
                fall back to -bios
            virt_flash_fdt()
                update FDT

You have László to thank for making me explain this in detail.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Acked-by: Laszlo Ersek <lersek@redhat.com>
Message-id: 20190416091348.26075-4-armbru@redhat.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agopflash_cfi01: New pflash_cfi01_legacy_drive()
Markus Armbruster [Tue, 7 May 2019 11:55:02 +0000 (12:55 +0100)] 
pflash_cfi01: New pflash_cfi01_legacy_drive()

Factored out of pc_system_firmware_init() so the next commit can reuse
it in hw/arm/virt.c.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190416091348.26075-3-armbru@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agopc: Rearrange pc_system_firmware_init()'s legacy -drive loop
Markus Armbruster [Tue, 7 May 2019 11:55:01 +0000 (12:55 +0100)] 
pc: Rearrange pc_system_firmware_init()'s legacy -drive loop

The loop does two things: map legacy -drive to properties, and collect
all the backends for use after the loop.  The next patch will factor
out the former for reuse in hw/arm/virt.c.  To make that easier,
rearrange the loop so it does the first thing first, and the second
thing second.

Signed-off-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Laszlo Ersek <lersek@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190416091348.26075-2-armbru@redhat.com
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/marcel/tags/rdma-pull-request' into staging
Peter Maydell [Tue, 7 May 2019 11:36:50 +0000 (12:36 +0100)] 
Merge remote-tracking branch 'remotes/marcel/tags/rdma-pull-request' into staging

RDMA queue

* pvrdma: Add support for SRQ

# gpg: Signature made Sat 04 May 2019 14:35:40 BST
# gpg:                using RSA key 36D4C0F0CF2FE46D
# gpg: Good signature from "Marcel Apfelbaum <marcel.apfelbaum@zoho.com>" [marginal]
# gpg:                 aka "Marcel Apfelbaum <marcel@redhat.com>" [marginal]
# gpg:                 aka "Marcel Apfelbaum <marcel.apfelbaum@gmail.com>" [marginal]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: B1C6 3A57 F92E 08F2 640F  31F5 36D4 C0F0 CF2F E46D

* remotes/marcel/tags/rdma-pull-request:
  hw/pvrdma: Add support for SRQ
  hw/rdma: Modify create/destroy QP to support SRQ
  hw/rdma: Add support for managing SRQ resource
  hw/rdma: Add SRQ support to backend layer

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoMerge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging
Peter Maydell [Tue, 7 May 2019 09:43:32 +0000 (10:43 +0100)] 
Merge remote-tracking branch 'remotes/thibault/tags/samuel-thibault' into staging

Update slirp submodule

To fix Windows on ARM.

# gpg: Signature made Sat 04 May 2019 13:40:33 BST
# gpg:                using RSA key E61DBB15D4172BDEC97E92D9DB550E89F0FA54F3
# gpg: Good signature from "Samuel Thibault <samuel.thibault@aquilenet.fr>" [unknown]
# gpg:                 aka "Samuel Thibault <sthibault@debian.org>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@gnu.org>" [unknown]
# gpg:                 aka "Samuel Thibault <samuel.thibault@inria.fr>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@labri.fr>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@ens-lyon.org>" [marginal]
# gpg:                 aka "Samuel Thibault <samuel.thibault@u-bordeaux.fr>" [unknown]
# gpg: WARNING: This key is not certified with sufficiently trusted signatures!
# gpg:          It is not certain that the signature belongs to the owner.
# Primary key fingerprint: 900C B024 B679 31D4 0F82  304B D017 8C76 7D06 9EE6
#      Subkey fingerprint: E61D BB15 D417 2BDE C97E  92D9 DB55 0E89 F0FA 54F3

* remotes/thibault/tags/samuel-thibault:
  Update slirp submodule

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2 years agoi2c-ddc: move it to hw/display
Paolo Bonzini [Mon, 25 Mar 2019 15:59:23 +0000 (16:59 +0100)] 
i2c-ddc: move it to hw/display

Move it together with the other EDID code.  hw/i2c should only
include the core and the adapters, not the slaves.

Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-id: 20190325155923.30987-1-pbonzini@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agoati-vga: Fix check for blt outside vram
BALATON Zoltan [Tue, 9 Apr 2019 10:56:18 +0000 (12:56 +0200)] 
ati-vga: Fix check for blt outside vram

Fix the check preventing calling pixman functions that would access
memory outside allocated vram. The r128 X driver sometimes seem to try
blits that span outside vram, this check prevents crashing QEMU in
that case. (The r128 X driver may have problems even on real hardware
so I'm not sure if it's a client bug or emulation problem but at least
QEMU should survive.)

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Tested-by: Andrew Randrianasulu <randrianasulu@gmail.com>
Message-Id: <20190409110732.5C5FF7465DB@zero.eik.bme.hu>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agoqxl: avoid unaligned pointer reads/writes
Daniel P. Berrangé [Fri, 12 Apr 2019 12:16:26 +0000 (13:16 +0100)] 
qxl: avoid unaligned pointer reads/writes

The SPICE_RING_PROD_ITEM() macro is initializing a local
'uint64_t *' variable to point to the 'el' field inside
the QXLReleaseRing struct. This uint64_t field is not
guaranteed aligned as the struct is packed.

Code should not take the address of fields within a
packed struct. Changing the SPICE_RING_PROD_ITEM()
macro to avoid taking the address of the field is
impractical. It is clearer to just remove the macro
and inline its functionality in the three call sites
that need it.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-Id: <20190412121626.19829-6-berrange@redhat.com>
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agovl: add -vga help support
Marc-André Lureau [Fri, 12 Apr 2019 15:27:13 +0000 (17:27 +0200)] 
vl: add -vga help support

Provide help output similar to other argument help handling:

$ qemu-system-x86_64 -vga help
none
std                  standard VGA (default)
cirrus               Cirrus VGA
vmware               VMWare SVGA
xenfb
qxl                  QXL VGA
virtio               Virtio VG

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190412152713.16018-3-marcandre.lureau@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agovl: constify VGAInterfaceInfo
Marc-André Lureau [Fri, 12 Apr 2019 15:27:12 +0000 (17:27 +0200)] 
vl: constify VGAInterfaceInfo

Signed-off-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Tested-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190412152713.16018-2-marcandre.lureau@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agohw/display/cirrus_vga: Remove unused include
Philippe Mathieu-Daudé [Sun, 5 May 2019 22:56:40 +0000 (00:56 +0200)] 
hw/display/cirrus_vga: Remove unused include

Commit ce3cf70edaaf split the ISA device out of the PCI one,
but forgot to remove the "hw/loader.h" header inclusion (the ISA
device calls rom_add_vga()).  Remove the now unused include.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Message-id: 20190505225640.4592-1-philmd@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agohw/display/cirrus_vga: Update the documentation URL
Philippe Mathieu-Daudé [Sat, 4 May 2019 12:16:50 +0000 (14:16 +0200)] 
hw/display/cirrus_vga: Update the documentation URL

The documentation URL is not working, but is backed up by the
Wayback Machine on the Internet Archive.
Replace the outdated link by a captured one.
Add another link to the VGADOC4b.ZIP archive content.

Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-id: 20190504121650.12651-1-philmd@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agoqxl: check release info object
Prasad J Pandit [Thu, 25 Apr 2019 06:35:34 +0000 (12:05 +0530)] 
qxl: check release info object

When releasing spice resources in release_resource() routine,
if release info object 'ext.info' is null, it leads to null
pointer dereference. Add check to avoid it.

Reported-by: Bugs SysSec <bugs-syssec@rub.de>
Signed-off-by: Prasad J Pandit <pjp@fedoraproject.org>
Message-id: 20190425063534.32747-1-ppandit@redhat.com
Signed-off-by: Gerd Hoffmann <kraxel@redhat.com>
2 years agodecodetree: Add DisasContext argument to !function expanders
Richard Henderson [Thu, 21 Mar 2019 02:21:31 +0000 (19:21 -0700)] 
decodetree: Add DisasContext argument to !function expanders

This does require adjusting all existing users.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agodecodetree: Expand a decode_load function
Richard Henderson [Thu, 31 Jan 2019 19:34:11 +0000 (11:34 -0800)] 
decodetree: Expand a decode_load function

Read the instruction, loading no more bytes than necessary.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
2 years agodecodetree: Initial support for variable-length ISAs
Richard Henderson [Thu, 31 Jan 2019 02:01:29 +0000 (18:01 -0800)] 
decodetree: Initial support for variable-length ISAs

Assuming that the ISA clearly describes how to determine
the length of the instruction, and the ISA has a reasonable
maximum instruction length, the input to the decoder can be
right-justified in an appropriate insn word.

This is not 100% convenient, as out-of-line %fields are
numbered relative to the maximum instruction length, but
this appears to still be usable.

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