Instead of manually parsing the boot_list as character stream,
we can access the nth boot device, specified by the position in the
boot order.
Signed-off-by: NDominik Dingel <dingel@linux.vnet.ibm.com>
Reviewed-by: NAnthony Liguori <aliguori@us.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Since we now have working firmware for s390-ccw in the tree, we can
default to it on our s390-ccw machine, rendering it more useful.
Signed-off-by: NAlexander Graf <agraf@suse.de>

Now that we have all the source code ready, add a compiled blob into
the QEMU source tree, so that people without access to an s390 compiler
can run the s390-ccw firmware.
Signed-off-by: NAlexander Graf <agraf@suse.de>

Ask the host about the configuration instead of guessing it.
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Try to handle at least some of the errors that may happen.
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

- Use tpi + tsch to get interrupts.
- Return an error if the irb indicates problems.
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

stsch is the canonical way to detect devices. As a bonus, we can
abort the loop if we get cc 3, and we need to check only the valid
devices (dnv set).
Signed-off-by: NCornelia Huck <cornelia.huck@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Lets fix this gcc warning:

virtio.c: In function ‘vring_send_buf’:
virtio.c:125:35: error: operation on ‘vr->next_idx’ may be undefined
[-Werror=sequence-point]
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

We have to call strip with s390-ccw.elf as input and
s390-ccw.img as output
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

Lets build the s390-ccw rom if on s390. Also fix the separate build
folder case.
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

dont waste cpu power on an error condition. Lets stop the guest
with a disabled wait.
Signed-off-by: NChristian Borntraeger <borntraeger@de.ibm.com>
Signed-off-by: NAlexander Graf <agraf@suse.de>

This patch adds a makefile, so we can build our ccw firmware. Also
add the resulting binaries to .gitignore, so that nobody is annoyed
they might be in the tree.
Signed-off-by: NAlexander Graf <agraf@suse.de>

On s390, there is an architected boot map format that we can read to
boot a certain entry off the disk. Implement a simple reader for this
that always boots the first (default) entry.
Signed-off-by: NAlexander Graf <agraf@suse.de>

Like all great programs, we have to call between different functions in
different object files. And all of them need a common ground of defines.

Provide a file that provides these defines.
Signed-off-by: NAlexander Graf <agraf@suse.de>

In order to boot, we need to be able to access a virtio-blk device through
the CCW bus. Implement support for this.
Signed-off-by: NAlexander Graf <agraf@suse.de>

In order to communicate with the user, we need an I/O mechanism that he
can read. Implement SCLP ASCII support, which happens to be the default
in the s390 ccw machine.

This file is missing read support for now. It can only print messages.
Signed-off-by: NAlexander Graf <agraf@suse.de>

This C file is the main driving piece of the s390 ccw firmware. It
provides a search for a workable block device, sets it as the default
to boot off of and boots from it.
Signed-off-by: NAlexander Graf <agraf@suse.de>

We want to write most of our code in C, so add a small assembly
stub that jumps straight into C code for us to continue booting.
Signed-off-by: NAlexander Graf <agraf@suse.de>

We have a virtio-s390 and a virtio-ccw machine in QEMU. Both use vastly
different ways to do I/O. Having the same firmware blob for both doesn't
really make any sense.

Instead, let's parametrize the firmware file name, so that we can have
different blobs for different machines.
Signed-off-by: NAlexander Graf <agraf@suse.de>

Our firmware blob is always a raw file that we load at a fixed address today.
Support loading an ELF blob instead that we can map high up in memory.

This way we don't have to be so conscious about size constraints.
Signed-off-by: NAlexander Graf <agraf@suse.de>

We can have different load addresses for different blobs we boot with.
Make the reset IP dynamic, so that we can handle things more flexibly.
Signed-off-by: NAlexander Graf <agraf@suse.de>

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

# By Liu Yuan (4) and others
# Via Stefan Hajnoczi
* stefanha/block:
  sheepdog: fix loadvm operation
  sheepdog: resend write requests when SD_RES_READONLY is received
  sheepdog: add helper function to reload inode
  sheepdog: add SD_RES_READONLY result code
  sheepdog: cleanup find_vdi_name
  rbd: Fix use after free in rbd_open()
  block: Disable driver-specific options for 1.5
  sheepdog: implement .bdrv_co_is_allocated()
  sheepdog: use BDRV_SECTOR_SIZE
  sheepdog: add discard/trim support for sheepdog
  block/ssh: Require libssh2 >= 1.2.8.

Message-id: 1366976682-10251-1-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

# By Ed Maste (2) and others
# Via Stefan Hajnoczi
* stefanha/trivial-patches:
  bsd-user: Track change in FreeBSD SYSCTL(9) types
  virtio: Fix compilation without CONFIG_VHOST_SCSI
  qemu-doc: Option -ignore-environment removed.
  s390x: use CONFIG_INT128 to detect __uint128_t
  linux-user: fix compile error due to stray colon at end of #ifdef line

Message-id: 1366975563-16216-1-git-send-email-stefanha@redhat.com
Signed-off-by: NAnthony Liguori <aliguori@us.ibm.com>

Currently the 'loadvm' opertaion works as following:
1. switch to the snapshot
2. mark current working VDI as a snapshot
3. rely on sd_create_branch to create a new working VDI based on the snapshot

This works not the same as other format as QCOW2. For e.g,

qemu > savevm # get a live snapshot snap1
qemu > savevm # snap2
qemu > loadvm 1 # This will steally create snap3 of the working VDI

Which will result in following snapshot chain:

base <-- snap1 <-- snap2 <-- snap3
          ^
          |
      working VDI

snap3 was unnecessarily created and might be annoying users.

This patch discard the unnecessary 'snap3' creation. and implement
rollback(loadvm) operation to the specified snapshot by
1. switch to the snapshot
2. delete working VDI
3. rely on sd_create_branch to create a new working VDI based on the snapshot

The snapshot chain for above example will be:

base <-- snap1 <-- snap2
          ^
          |
      working VDI

Cc: qemu-devel@nongnu.org
Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NLiu Yuan <tailai.ly@taobao.com>
Reviewed-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

When a snapshot is taken from out side of qemu (e.g. qemu-img
snapshot), write requests to the current vdi return SD_RES_READONLY.
In this case, the sheepdog block driver needs to update the current
inode to the latest one and resend the write requests.
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

This adds a helper function to update the current inode state with the
specified vdi object.
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Sheepdog returns SD_RES_READONLY when qemu sends write requests to the
snapshot vdi.  This adds the result code and makes sd_strerror() print
its error reason.
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

This makes 'filename' and 'tag' constant variables, and renames
'for_snapshot' to 'lock' to clear how it works.
Signed-off-by: NMORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Commit a9ccedc3 frees the QemuOpts for the driver-specific options
immediately, even though it still needs the filename string that is
contained there. This doesn't work. Move the deletion of the QemuOpts to
the end of the function where its content isn't needed any more.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

We don't want to commit to the API yet before everything is worked out.
Disable it for the 1.5 release. This commit is meant to be reverted
after the 1.5 release.

The disabling of the driver-specific options is achieved by applying the
old checks while parsing the command line.
Signed-off-by: NKevin Wolf <kwolf@redhat.com>
Reviewed-by: NEric Blake <eblake@redhat.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NLiu Yuan <tailai.ly@taobao.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Signed-off-by: NLiu Yuan <tailai.ly@taobao.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

The 'TRIM' command from VM that is to release underlying data storage for
better thin-provision is already supported by the Sheepdog.

This patch adds the TRIM support at QEMU part.

For older Sheepdog that doesn't support it, we return 0(success) to upper layer.

Cc: MORITA Kazutaka <morita.kazutaka@lab.ntt.co.jp>
Cc: Kevin Wolf <kwolf@redhat.com>
Cc: Stefan Hajnoczi <stefanha@redhat.com>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: NLiu Yuan <tailai.ly@taobao.com>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

libssh2 >= 1.2.8 is required to enable this block device (because
that version introduced the libssh2_session_handshake call).

Change the test to use pkg-config exclusively.  If the user requests
--enable-libssh2 and the minimum version is not available, then the
following error is displayed:

  $ ./configure --enable-libssh2

  ERROR: libssh2 >= 1.2.8 required for --enable-libssh2

If --enable-libssh2 is not specified, then the feature is silently
disabled if sufficiently new libssh2 is not available.
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Originally from Garrett Cooper in FreeBSD PR ports/155558
http://www.freebsd.org/cgi/query-pr.cgi?pr=155558Signed-off-by: NEd Maste <emaste@freebsd.org>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Signed-off-by: NEd Maste <emaste@freebsd.org>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Has been removed in commit fc9c5412.
Signed-off-by: NThomas Schwinge <thomas@codesourcery.com>
Reviewed-by: NPeter Maydell <peter.maydell@linaro.org>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Target s390x uses ad-hoc macro magic to guess if the compiler
supports the GCC extension __uint128_t.  This patch uses the
the dedicated macro CONFIG_INT128 defined by configure instead.

This fixes compilation with the CIL source code analyzer, which
uses GCC as a preprocessor but does not support __uint128_t.
Signed-off-by: NGabriel Kerneis <gabriel@kerneis.info>
Reviewed-by: NRichard Henderson <rth@twiddle.net>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>

Remove a stray colon from the end of a #ifdef line. Some versions
of gcc complain about this:
 linux-user/syscall.c: In function ‘do_syscall’:
 linux-user/syscall.c:7606:28: error: extra tokens at end of #ifdef directive [-Werror]
Signed-off-by: NPeter Maydell <peter.maydell@linaro.org>
Acked-By: NRiku Voipio <riku.voipio@linaro.org>
Signed-off-by: NStefan Hajnoczi <stefanha@redhat.com>