We create optional sections with this patch.  But we already have
optional subsections.  Instead of having two mechanism that do the
same, we can just generalize it.

For subsections we just change:

- Add a needed function to VMStateDescription
- Remove VMStateSubsection (after removal of the needed function
  it is just a VMStateDescription)
- Adjust the whole tree, moving the needed function to the corresponding
  VMStateDescription
Signed-off-by: NJuan Quintela <quintela@redhat.com>

When setting "realized" fails, scsi_bus_legacy_add_drive() passes the
error to qerror_report_err(), then returns an unspecific "Setting
drive property failed" error, which is reported further up the call
chain.

Example:

    $ qemu-system-x86_64 -nodefaults -S -display none \
    > -drive if=scsi,id=foo,file=tmp.qcow2 -global isa-fdc.driveA=foo
    qemu-system-x86_64: -drive if=scsi,id=foo,file=tmp.qcow2: Property 'scsi-disk.drive' can't take value 'foo', it's in use
    qemu-system-x86_64: Setting drive property failed
    qemu-system-x86_64: Initialization of device lsi53c895a failed: Device initialization failed

Clean up the obvious way: simply return the original error to the
caller.  Gets rid of the second message in the above error cascade.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-Id: <1425925048-15482-4-git-send-email-armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Three kinds of callers:

1. On failure, report the error and abort

   Passing &error_abort does the job.  No functional change.

2. On failure, report the error and exit()

   This is qdev_prop_set_drive_nofail().  Error reporting moves from
   qdev_prop_set_drive() to its caller.  Because hiding away the error
   in the monitor right before exit() isn't helpful, replace
   qerror_report_err() by error_report_err().  Shouldn't make a
   difference, because qdev_prop_set_drive_nofail() should never be
   used in QMP context.

3. On failure, report the error and recover

   This is usb_msd_init() and scsi_bus_legacy_add_drive().  Error
   reporting and freeing the error object moves from
   qdev_prop_set_drive() to its callers.

   Because usb_msd_init() can't run in QMP context, replace
   qerror_report_err() by error_report_err() there.

   No functional change.

   scsi_bus_legacy_add_drive() calling qerror_report_err() is of
   course inappropriate, but this commit merely makes it more obvious.
   The next one will clean it up.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-Id: <1425925048-15482-3-git-send-email-armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Commit a818a4b6 changed scsi_bus_legacy_handle_cmdline() to report
errors from scsi_bus_legacy_add_drive() with error_report() in
addition to returning them.  That's inappropriate.

Two kinds of callers:

1. realize methods (devices "esp", "virtio-scsi-device" and
   "spapr-vscsi")

   The error object gets passed up the call chain until it gets
   reported again and freed.

   Example:

   $ qemu-system-arm -M virt -S -display none \
   > -drive if=scsi,id=foo,bus=1,file=tmp.qcow2 \
   > -device nec-usb-xhci -device usb-storage,drive=foo \
   > -device virtio-scsi-pci
   qemu-system-arm: -drive if=scsi,id=foo,bus=1,file=tmp.qcow2: Property 'scsi-disk.drive' can't take value 'foo', it's in use
   qemu-system-arm: -drive if=scsi,id=foo,bus=1,file=tmp.qcow2: Setting drive property failed
   qemu-system-arm: -device virtio-scsi-pci: Setting drive property failed
   qemu-system-arm: -device virtio-scsi-pci: Device initialization failed
   qemu-system-arm: -device virtio-scsi-pci: Device 'virtio-scsi-pci' could not be initialized

   The second message in this error cascade comes from
   scsi_bus_legacy_handle_cmdline().  The error object then gets
   passed up to the qdev_init() called from
   virtio_scsi_pci_init_pci(), which reports it again.

2. init methods (devices "am53c974", "dc390", "lsi53c895a",
   "lsi53c810", "megasas", "megasas-gen2")

   init methods need to report their errors with qerror_report().
   These don't.  The inappropriate error_report() papers over the bug.

   error_report() isn't the same as qerror_report() in QMP context,
   but this can't actually happen: QMP can still only hot-plug, and
   callers call scsi_bus_legacy_handle_cmdline() only on cold-plug.
   Except for sysbus_esp_realize(), but that can't be hot-plugged at
   all, as far as I can tell.

Fix the init methods and drop the inappropriate error_report() in
scsi_bus_legacy_handle_cmdline().
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NPeter Crosthwaite <peter.crosthwaite@xilinx.com>
Message-Id: <1425925048-15482-2-git-send-email-armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

This mimics what is done in qdev_device_add, and lets the device be
freed in case something goes wrong.  Otherwise, object_unparent returns
immediately without freeing the device, which is on the other hand left
in the parent bus's list of children.

scsi_bus_legacy_handle_cmdline then returns an error, and the HBA is
destroyed as well with object_unparent.  But the lingering device that
was not removed in scsi_bus_legacy_add_drive cannot be removed now either,
and bus_unparent gets stuck in an infinite loop trying to empty the list
of children.

The right fix of course would be to assert in bus_add_child that the
device already has a bus, and remove the "safety net" that adds the
drive to the QOM tree in device_set_realized.  I am not yet sure whether
that would entail changing all callers to qdev_create (as well as
isa_create and usb_create and the corresponding _try_create versions).
Reported-by: NMarkus Armbruster <armbru@redhat.com>
Tested-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

scsi_req_cancel_complete is responsible for releasing the request, so we
shouldn't skip it in any case. This doesn't affect the only existing
caller, virtio-scsi, but is useful for other devices once they use it.
Suggested-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Commit d5776465 (scsi: Introduce scsi_req_cancel_complete, 2014-09-25)
was supposed to have no semantic change, but it missed a case.  When
r->aiocb has already been NULLed, but DMA was not complete and the
SCSI layer was waiting for scsi_req_continue, after the patch the
SCSI layer will not call the .cancel callback of SCSIBusInfo.

Fixes: d5776465
Cc: qemu-stable@nongnu.org
Reported-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Tested-by: NDr. David Alan Gilbert <dgilbert@redhat.com>
Reviewed-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

All implementations are the same.
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The EFI firmware doesn't handle unit attentions properly,
so we need to clear the Power On/Reset unit attention upon
initial reset.
Signed-off-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

scsi_cdb_length() does not return the length of the cdb, but
the transfersize encoded in the cdb. So rename it to scsi_cdb_xfer()
and also rename all other related functions to end with _xfer.

We can then add a new scsi_cdb_length() which actually does return the
length of the cdb.  With that DEBUG_SCSI can now display the correct
CDB buffer.
Signed-off-by: NHannes Reinecke <hare@suse.de>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Device models should access their block backends only through the
block-backend.h API.  Convert them, and drop direct includes of
inappropriate headers.

Just four uses of BlockDriverState are left:

* The Xen paravirtual block device backend (xen_disk.c) opens images
  itself when set up via xenbus, bypassing blockdev.c.  I figure it
  should go through qmp_blockdev_add() instead.

* Device model "usb-storage" prompts for keys.  No other device model
  does, and this one probably shouldn't do it, either.

* ide_issue_trim_cb() uses bdrv_aio_discard() instead of
  blk_aio_discard() because it fishes its backend out of a BlockAIOCB,
  which has only the BlockDriverState.

* PC87312State has an unused BlockDriverState[] member.

The next two commits take care of the latter two.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NMax Reitz <mreitz@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

The patch is big, but all it really does is replacing

    dinfo->bdrv

by

    blk_bs(blk_by_legacy_dinfo(dinfo))

The replacement is repetitive, but the conversion of device models to
BlockBackend is imminent, and will shorten it to just
blk_legacy_dinfo(dinfo).

Line wrapping muddies the waters a bit.  I also omit tests whether
dinfo->bdrv is null, because it never is.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Reviewed-by: NBenoît Canet <benoit.canet@nodalink.com>
Reviewed-by: NMax Reitz <mreitz@redhat.com>
Reviewed-by: NKevin Wolf <kwolf@redhat.com>
Signed-off-by: NKevin Wolf <kwolf@redhat.com>

Remove bootindex form qdev property to qom, things will
continue to work just fine, and we can use qom features
which are not supported by qdev property.
Signed-off-by: NGonglei <arei.gonglei@huawei.com>
Reviewed-by: NGerd Hoffmann <kraxel@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

Add a qom property with the same name 'bootindex',
when we remove it form qdev property, things will
continue to work just fine, and we can use qom features
which are not supported by qdev property.
Signed-off-by: NGonglei <arei.gonglei@huawei.com>
Reviewed-by: NGerd Hoffmann <kraxel@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

SCSI subsytem was converted to hotplug handler API and
doesn't use SCSIBusInfo{hotplug, hot_unplug} fields and
related callbacks anymore.
Signed-off-by: NIgor Mammedov <imammedo@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAndreas Färber <afaerber@suse.de>

That would allow to handle SCSI device unplug
on HBAs without dedicated hot(un)plug handlers
and avoid making such HBAs explicitly hotpluggable.
Signed-off-by: NIgor Mammedov <imammedo@redhat.com>
Reviewed-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NAndreas Färber <afaerber@suse.de>

Devices will call this function to start an asynchronous cancellation. The
bus->info->cancel will be called after the request is canceled.

Devices will probably need to track a separate TMF request that triggers this
cancellation, and wait until the cancellation is done before completing it. So
we store a notifier list in SCSIRequest and in scsi_req_cancel_complete we
notify them.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Let the aio cb do the clean up and notification job after scsi_req_cancel, in
preparation for asynchronous cancellation.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The only two implementations are identical to each other, with nothing specific
to device: they only call bdrv_aio_cancel with the SCSIRequest.aiocb.

Let's move it to scsi-bus.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The only user of this function is spapr_vscsi.c. We can convert to
scsi_req_cancel plus adding a check in vscsi_request_cancelled.
Suggested-by: NPaolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NFam Zheng <famz@redhat.com>
[Drop prototype. - Paolo]
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Zeroing sense buffer for each scsi request is not efficient, we can just
leave it uninitialized because sense_len is set to 0.

Move the implicitly zeroed fields to the end of the structure and use a
partial memset.

The explicitly initialized fields (by scsi_req_alloc or scsi_req_new)
are moved to the beginning of the structure, before sense buffer, to
skip the memset.

Also change g_malloc0 to g_slice_alloc.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Replace "init/destroy" with "realize/unrealize" in SCSIDeviceClass,
which has errp as a parameter. So all the implementations now use
error_setg instead of error_report for reporting error.

Also in scsi_bus_legacy_handle_cmdline, report the error when
initializing the if=scsi devices, before returning it, because in the
callee, error_report is changed to error_setg. And the callers don't
have the right locations (e.g. "-drive if=scsi").
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The callback lets the bus provide the direction and transfer count
for passthrough commands, enabling passthrough of vendor-specific
commands.
Reviewed-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

These callbacks will let devices do their own request parsing, or
defer it to the bus.  If the bus does not provide an implementation,
in turn, fall back to the default parsing routine.

Swap the first two arguments to scsi_req_parse, and rename it to
scsi_req_parse_cdb, for consistency.
Reviewed-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The per-SCSIDevice parse_cdb callback must not be called if the
request will go through special SCSIReqOps, so detect the special
cases early enough.
Reviewed-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

In the case that the lun number is taken by another scsi device, don't
release the existing device siliently, but report an error to user.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

This makes scsi_command_name() public.

This makes use of scsi_command_name() in debug output for scsi-disk and
spapr-vscsi host bus adapter. Before this, SCSI used to print hex numbers
instead of human-friendly strings.

This adds GET_EVENT_STATUS_NOTIFICATION and READ_DISC_INFORMATION to
the list of SCSI commands supported by scsi_command_name().
Signed-off-by: NAlexey Kardashevskiy <aik@ozlabs.ru>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

After previous Peter patch, they are redundant.  This way we don't
assign them except when needed.  Once there, there were lots of case
where the ".fields" indentation was wrong:

     .fields = (VMStateField []) {
and
     .fields =      (VMStateField []) {

Change all the combinations to:

     .fields = (VMStateField[]){

The biggest problem (appart from aesthetics) was that checkpatch complained
when we copy&pasted the code from one place to another.
Signed-off-by: NJuan Quintela <quintela@redhat.com>
Acked-by: NAlexey Kardashevskiy <aik@ozlabs.ru>
Reviewed-by: NMichael S. Tsirkin <mst@redhat.com>

For clarity, and to hush up Coverity.
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

This fixes warnings from the static code analysis (smatch).
Signed-off-by: NStefan Weil <sw@weilnetz.de>
Signed-off-by: NMichael Tokarev <mjt@tls.msk.ru>

This assertion is invalid, because get_sg_list can return an
empty sg-list even for commands that transfer no data (such
as SYNCHRONIZE CACHE).
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

c5f52875 changed the size of sense array in vmstate_scsi_device by
mistake. This patch restores the old size, and add a subsection for the
remaining part of the buffer size. So that migration is not broken.
Signed-off-by: NFam Zheng <famz@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The transfer length depends on field BYTCHK, which is encoded in byte
1, bits 1..2.  However, the guard for for case BYTCHK=11b doesn't
work, and we get case 01b instead.  Fix it.

Note that since emulated scsi-hd fails the command outright, it takes
SCSI passthrough of a device that actually implements VERIFY with
BYTCHK=11b to make the bug bite.

Screwed up in commit d12ad44c.  Spotted by Coverity.

Cc: qemu-stable@nongnu.org
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

SCSI defines a status code for when a thin-provisioned LUNs would
exceed the allocated space, map ENOSPC to it.
Reviewed-by: NStefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

SeaBIOS waits for LUN0 to respond to the TEST UNIT READY command
in order to decide whether it should part of the boot sequence.
If LUN0 does not respond to the command, boot is delayed by up
to 5 seconds.  This currently happens when there is no LUN0 on
a target.  Fix that by adding a trivial implementation of the
command.

Cc: qemu-stable@nongnu.org
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

The amount of bytes to transfer depends on the BYTCHK field.
If any data is transferred, it is sent to the device.

Cc: qemu-stable@nongnu.org
Tested-by: NHervé Poussineau <hpoussin@reactos.org>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

Signed-off-by: NHans de Goede <hdegoede@redhat.com>
Signed-off-by: NGerd Hoffmann <kraxel@redhat.com>

The qdev_free() function name is misleading since all the function does
is unlink the device from its parent.  The device is not necessarily
freed.

The device will be freed when its QObject refcount reaches zero.  It is
usual for the parent (bus) to hold the final reference but there are
cases where something else holds a reference so "free" is a misleading
name.

Call object_unparent(obj) directly instead of having a qdev wrapper
function.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NAndreas Färber <afaerber@suse.de>

r->buf is hardcoded to 2056 which is (256 + 1) * 8, allowing 256 luns at
most. If more than 256 luns are specified by user, we have buffer
overflow in scsi_target_emulate_report_luns.

To fix, we allocate the buffer dynamically.
Signed-off-by: NAsias He <asias@redhat.com>
Tested-by: NMichael Roth <mdroth@linux.vnet.ibm.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>

scsi_bus_legacy_add_drive() creates either a scsi-disk or a
scsi-generic device.  It sets property "serial" to argument serial
unless null.  Crashes with scsi-generic, because it doesn't have such
the property.

Only usb_msd_initfn_storage() passes non-null serial.  Reproducer:

    $ qemu-system-x86_64 -nodefaults -display none -S -usb \
    -drive if=none,file=/dev/sg1,id=usb-drv0 \
    -device usb-storage,id=usb-msd0,drive=usb-drv0,serial=123
    qemu-system-x86_64: -device usb-storage,id=usb-msd0,drive=usb-drv0,serial=123: Property '.serial' not found
    Aborted (core dumped)

Fix by handling exactly like "removable": set the property only when
it exists.

Cc: qemu-stable@nongnu.org
Reviewed-by: NAndreas Färber <afaerber@suse.de>
Signed-off-by: NMarkus Armbruster <armbru@redhat.com>
Signed-off-by: NPaolo Bonzini <pbonzini@redhat.com>