Return the correct value in the domain field in the cp15 DFSR
(C5) -- bug noticed during Xvisor development.
Signed-off-by: NJean-Christophe DUBOIS <jcd@tribudubois.net>
[Peter Maydell: reworded commit message]
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Destroying a mutex that another thread might have just unlocked
is racy.  It usually works, but you cannot do that in general and
can lead to deadlocks or segfaults.  Change ccid to use joinable
threads instead.

(Also, qemu_mutex_init/qemu_cond_init were missing).
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Rewrite the handshaking between qemu_thread_create and the
win32_start_routine, so that the thread can be joined without races.
Similar handshaking is done now between qemu_thread_exit and
qemu_thread_join.

This also simplifies how QemuThreads are initialized.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Allow to control if a QEMU thread is created joinable or not. Make it
not joinable by default to avoid that we keep the associated resources
around when terminating a thread without joining it (what we couldn't do
so far for obvious reasons).

The audio subsystem will need the join feature when converting it to
QEMU threading/locking abstractions, so provide that service.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Split from Jan's original qemu-thread-posix.c patch.  No semantic change,
just introduce the new API that POSIX and Win32 implementations will
conform to.
Signed-off-by: NJan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

The Symbian Virtual Platform was an ARM-based development and debugging
board.  Since Symbian has been disbanded and the code is no longer being
used it can now be removed.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Today net/socket.c has no consistent policy for closing the socket file
descriptor when initialization fails.  This means we leak the file
descriptor in some cases or we could also try to close it twice.

Make error paths consistent by taking ownership of the file descriptor
and closing it on error.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

In order to make later patches sane, expand the tab characters and
conform to QEMU coding style now.
Signed-off-by: NStefan Hajnoczi <stefanha@linux.vnet.ibm.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Not that there is blacklisting functionality we can no longer infer
the agent's capabilities via version. This patch extends the current
guest-info RPC to also return a list of dictionaries containing the name
of each supported RPC, along with a boolean indicating whether or not
the command has been disabled by a guest administrator/distro.
Signed-off-by: NMichael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This adds a command-line option, -b/--blacklist, that accepts a
comma-seperated list of RPCs to disable, or prints a list of
available RPCs if passed "?".

In consequence this also adds general blacklisting and RPC listing
facilities to the new QMP dispatch/registry facilities, should the
QMP monitor ever have a need for such a thing.

Ideally, to avoid support/compatability issues in the future,
blacklisting guest agent functionality will be the exceptional
case, but we add the functionality here to handle guest administrators
with specific requirements.
Signed-off-by: NMichael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: NMichael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Add a .mailmap file so 'git shortlog' can map the unfriendly
pre-git-conversion author entries to real names.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

In case close() fails, we want to report the error back.

Changes v1 -> v2:
 - Use braces on if statement to match coding style
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

In case close() fails, we want to report the error back.

Changes v1 -> v2:
 - Use braces on if statement to match coding style
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

All qemu_fclose() callers were already changed to accept any negative
value as error, so we now can change it to return -errno.

When the process exits with a non-zero exit code, we return -EIO to as a
fake errno value.

Changes v1 -> v2:
 - Don't use "//" comments, to make checkpatch.pl happy
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This is what qemu_fclose() expects.

Changes v1 -> v2:
 - Add braces to if statement to match coding style
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This is what qemu_fclose() expects.

Changes v1 -> v2:
 - On success, keep returning pclose() return value, instead of always 0.

Changes v2 -> v3:
 - Add braces on if statements to match coding style
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This will make sure no error will be missed as long as callers always
check for qemu_fclose() return value. For reference, this is the
complete list of qemu_fclose() callers:

 - exec_close(): already fixed to check for negative values, not -1
 - migrate_fd_cleanup(): already fixed to consider only negative values
   as error, not any non-zero value
 - exec_accept_incoming_migration(): no return value check (yet)
 - fd_accept_incoming_migration(): no return value check (yet)
 - tcp_accept_incoming_migration(): no return value check (yet)
 - unix_accept_incoming_migration(): no return value check (yet)
 - do_savevm(): no return value check (yet)
 - load_vmstate(): no return value check (yet)

Changes v1 -> v2:
 - Add small comment about the need to return previously-spotted errors

Changes v2 -> v3:
 - Add braces to "if" statements to match coding style
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Also, we now return the qemu_fclose() value unchanged to the caller. For
reference, the migrate_fd_cleanup() callers are the following:

- migrate_fd_completed(): any negative value is considered an
  error, so the change is OK.
- migrate_fd_error(): doesn't check the migrate_fd_cleanup() return value
- migrate_fd_cancel(): doesn't check the migrate_fd_cleanup() return
  value
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Note that we don't return the unchanged return value back yet, because
we need to change all qemu_fclose() callers to accept any positive value
as success.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

qemu_fclose() and QEMUFile->close will return -errno on error, and any
positive value on success.

We need the positive non-zero success values because
migration-exec.c:exec_close() relies on non-zero return values to get
the process exit code.

Changes v1 -> v2:
 - Cosmetic spelling change on comment text
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Some code uses qemu_file_set_error() already, so use it everywhere
when setting last_error, for consistency.
Signed-off-by: NEduardo Habkost <ehabkost@redhat.com>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

"!X == 2" is always false (spotted by Coverity), so the checks
for whether rndis is in the correct state would never fire.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the usb-uhci device,
introduced by commit fff23ee9
'usb-uhci: Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the usb-ehci device,
introduced by commit 68d55358
'usb-ehci: Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the rtl8139 device,
introduced by commit 3ada003a
'rtl8139: Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the lsi53c895a device,
introduced by commit 9ba4524c
'lsi53c895a: Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the PCI IDE device,
introduced by commit 552908fe
'PCI IDE: Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the e1000 device,
introduced by commit 62ecbd35 'e1000:
Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This patch removes some unnecessary casts in the eepro100 device,
introduced by commit 16ef60c9
'eepro100: Use PCI DMA stub functions'.
Signed-off-by: NDavid Gibson <david@gibson.dropbear.id.au>
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NStefan Sandstrom <Stefan.Sandstrom@axis.com>
Signed-off-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>

hw/mpcore.c is now implementing only ARM11MPCore specific peripherals,
and is #included only from hw/arm11mpcore.c, so just merge it into that
file.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Implement the A9MP private peripheral region correctly, rather
than piggybacking on the 11MPCore code; the two CPUs are not the
same in this area.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

The only code left in mpcore_priv_read and mpcore_priv_write is now
the implementation of the SCU registers. Clean up by renaming functions
and removing some unnecessary conditionals to make this clearer.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Use the GIC provided memory region for the CPU interface rather
than implementing our own.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Switch to using the GIC memory regions for the CPU interface
rather than hand implementing them as a subcase of mpcore_priv_read()
and mpcore_priv_write().
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Expose the ARM GIC CPU interfaces as memory regions, rather than
just providing read and write functions for them.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Turn the ARM MPcore private timer/watchdog blocks into separate
qdev devices. This will allow us to share them neatly between
11MPCore and A9MPcore.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>

Properly implement dual-timer read/write for the sp804 dual timer module.
Based on ARM specs at
http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0271d/index.htmlSigned-off-by: NHans Jang <hsjang@ok-labs.com>
Signed-off-by: NDavid Mirabito <david.mirabito@nicta.com.au>
Signed-off-by: NPeter Chubb <peter.chubb@nicta.com.au>
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>