Adapt the signature of x86_cpu_realize(), hook up to
DeviceClass::realize and set realized = true in cpu_x86_init().

The QOM realizefn cannot depend on errp being non-NULL as in
cpu_x86_init(), so use a local Error to preserve error handling behavior
on APIC initialization errors.
Reviewed-by: NIgor Mammedov <imammedo@redhat.com>
Reviewed-by: NEduardo Habkost <ehabkost@redhat.com>
[AF: Invoke parent's realizefn]
Signed-off-by: NAndreas Färber <afaerber@suse.de>

Turn arm_cpu_realize() into a QOM realize function, no longer called
via cpu.h prototype. To maintain the semantics of cpu_init(), set
realized = true explicitly in cpu_arm_init().

Move GDB coprocessor registration, CPU reset and vCPU initialization
into the realizefn.
Signed-off-by: NAndreas Färber <afaerber@suse.de>

Update the alpha_cpu_realize() signature and hook up to
DeviceClass::realize. Set realized = true in cpu_alpha_init().

qapi/error.h is included through qdev now and no longer needed.
Acked-by: NRichard Henderson <rth@twiddle.net>
[AF: Invoke parent's realizefn]
Signed-off-by: NAndreas Färber <afaerber@suse.de>

Overwrite the default implementation with a no-op, no longer
attempting to call DeviceClass::init.
Signed-off-by: NAndreas Färber <afaerber@suse.de>

Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

The is the only remaining user.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

This is the only remaining user.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Use clz32 directly.  Which makes slightly more sense given
that the input is type "int" and not type "long".
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

A memory size of zero is invalid, and so that edge condition
does not occur.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

The use of ctz has already eliminated zero, and thus the difference
in edge conditions between the two routines is irrelevant.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Both uses of ctz have already eliminated zero, and thus the difference
in edge conditions between the two routines is irrelevant.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Acked-by: NPaolo Bonzini <pbonzini@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Add function comments to the routines, documenting the corner
cases upon which we are standardizing.  Fix the few instances
of non-standard coding style.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

We will standardize on these names, rather than the similar routines
currently residing in qemu/bitops.h.
Signed-off-by: NRichard Henderson <rth@twiddle.net>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Rename the typedef CPULogItem and the public array cpu_log_items
to names that better reflect the fact that the qemu_log functionality
isn't restricted to TCG CPU debug logs any more.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

The set_cpu_log() function in cpus.c is a fairly simple wrapper
which is only called from one location. Just inline the code
into vl.c, since there is no need to indirect it via cpus.c
and the handling of the error case is more appropriate to vl.c.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Rename the public-facing function cpu_set_log to qemu_set_log. This
requires us to rename the internal-only qemu_set_log() to
do_qemu_set_log().
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Rename cpu_str_to_log_mask() to qemu_str_to_log_mask(), since
the qemu_log functionality is no longer restricted to TCG CPU
debug logging.
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Abstract out the "print a human readable list of all the
valid log categories" functionality which is currently duplicated
in three separate places. (We leave the monitor.c help_cmd()
implementation as-is since it wants to send the message to
the monitor and add its own information.)
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

The qemu_log() functionality is no longer specific to TCG CPU debug logs.
Rename cpu_set_log_filename() to qemu_set_log_filename() and drop the
pointless wrapper set_cpu_log_filename().
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

It's worth to clean-up translation blocks variables and move them
into one context as was suggested by Swirl.
Also if we use this context directly inside tcg_ctx, then it
speeds up code generation a bit.
Signed-off-by: NEvgeny Voevodin <evgenyvoevodin@gmail.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Signed-off-by: NEvgeny Voevodin <evgenyvoevodin@gmail.com>
Signed-off-by: NBlue Swirl <blauwirbel@gmail.com>

Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

This reverts commit 10442558.

With the updated OpenBIOS image, -M g3beige fails to boot quik.
Signed-off-by: NAlexander Graf <agraf@suse.de>
Signed-off-by: NMark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Message-id: 1360796292-27078-1-git-send-email-agraf@suse.de
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

The device returns false from the can receive function when the FIFO is
full. This means the device should check for buffered input whenever a byte is
popped from the FIFO.
Reported-by: NJason Wu <huanyu@xilinx.com>
Signed-off-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-id: 1360632571-25638-1-git-send-email-peter.crosthwaite@xilinx.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

The curl_easy_setopt(state->curl, CURLOPT_PROTOCOLS, ...) interface was
introduced in libcurl 7.19.4.  Therefore we cannot protect against
CVE-2013-0249 when linking against an older libcurl.

This fixes the build failure introduced by
fb6d1bbd.
Reported-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Tested-by: NAndreas Färber <andreas.faeber@web.de>
Message-id: 1360743934-8337-1-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Simplifies the schema and the code.

QMP command

    { "execute" : "chardev-add",
      "arguments" : { "id" : "ser0",
                      "backend" : { "type" : "port",
                                    "data" : { "type": "serial",
                                               "device":"/dev/ttyS0"} } } }

becomes

    { "execute" : "chardev-add",
      "arguments" : { "id" : "ser0",
                      "backend" : { "type" : "serial",
                                    "data" : { "device":"/dev/ttyS0"} } } }

Bonus: nicer error messages.  "unknown chardev port (1)" becomes
"character device backend type 'parallel' not supported".
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>
Message-id: 1360767256-610-3-git-send-email-armbru@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Broken in commit d59044ef.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NLaszlo Ersek <lersek@redhat.com>
Message-id: 1360767256-610-2-git-send-email-armbru@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

The pSeries machine and some other devices don't supply a cleanup
callback. Revert part of 1ceef9f2 that
started calling it unconditionally.

Cc: Jason Wang <jasowang@redhat.com>
Signed-off-by: NAndreas Färber <afaerber@suse.de>
Message-id: 1360707366-9271-1-git-send-email-afaerber@suse.de
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

There are two ways to express an interruption subclass:
- As a bitmask, as used in cr6.
- As a number, as used in the I/O interruption word.

Unfortunately, we have treated to I/O interruption word as if it
contained the bitmask as well, which went unnoticed so far as
- (queued-for-next) kvm made the same mistake, and
- Linux guest kernels don't check the isc value in the I/O interruption
  word for subchannel interrupts.

Make sure that we treat the I/O interruption word correctly.
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

do_io_interrupt() would stop scanning further iscs if it found
an I/O interrupt it could inject. This might cause the pending
interrupt indication for I/O interrupts to be reset although there
might be queued I/O interrupts for subsequent iscs.

Fix this by reordering the logic: Inject the I/O interrupt immediately
and continue searching all iscs for queued interrupts.
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Starting a qemu with an sclp console and pressing a key very early
can result in
"qemu-system-s390x: hw/s390x/sclpconsole.c:60: receive_from_chr_layer:
Assertion `scon->iov' failed."

Lets make sure that the init process is finished, since the iov is
allocated after CHR_EVENT_OPENED by also checking for scon->iov.
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This function has nothing to do with ethernet. Its reusable for all DMA clients.
Dropped the "ethernet" in the name accordingly.
Signed-off-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Signed-off-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>

Pulled the qdev_create functionality out of xilinx_axiethernet_create() and
pushed it up to the petalogix_ml605_mmu machine model. This makes the ethernet
create+init process consistent with the AXI DMA. Renamed function to
xilinx_axiethernet_init accordingly.
Signed-off-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Signed-off-by: NEdgar E. Iglesias <edgar.iglesias@gmail.com>

* stefanha/block:
  Revert "block/vpc: Fix size calculation"
  block/raw-posix: detect readonly Linux block devices using BLKROGET

The return value of .save_live_pending() is the number of bytes
remaining.  This is just an estimate because we do not know how many
blocks will be dirtied by the running guest.

Currently our return value for .save_live_pending() is wrong because it
includes dirty blocks but not in-flight bdrv_aio_readv() requests or
unsent blocks.  Crucially, it also doesn't include the bulk phase where
the entire device is transferred - therefore we risk completing block
migration before all blocks have been transferred!

The return value of .save_live_iterate() is the number of bytes
transferred this iteration.  Currently we return whether there are bytes
remaining, which is incorrect.

Move the bytes remaining calculation into .save_live_pending() and
really return the number of bytes transferred this iteration in
.save_live_iterate().

Also fix the %ld format specifier which was used for a uint64_t
argument.  PRIu64 must be use to avoid warnings on 32-bit hosts.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Message-id: 1360661835-28663-3-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Migration .save_live_iterate() functions return the number of bytes
transferred.  The easiest way of doing this is by calling qemu_ftell(f)
at the beginning and end of the function to calculate the difference.

Make qemu_ftell() public so that block-migration will be able to use it.
Also adjust the ftell calculation for writable files where buf_offset
does not include buf_size.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NJuan Quintela <quintela@redhat.com>
Message-id: 1360661835-28663-2-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

g_thread_create() was deprecated in favor of g_thread_new() and
g_cond_new() was deprecated in favor of GCond initialization.  If the
host has glib 2.31 or newer, avoid using the deprecated functions.

This patch solves compiler warnings that are generated when glib's
deprecated functions are used.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Message-id: 1360676045-9204-3-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>