m_print() used special format specifiers from BSD's printf(9).

http://www.unix.com/man-page/freebsd/9/printf/



Instead of re-implementing these specifiers in different ways and
because this method is not being used anywhere, it was completely
removed.

Signed-off-by: Or Cohen <orc@fewbytes.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

ether_sprintf() used special format specifiers from BSD's printf(9).

http://www.unix.com/man-page/freebsd/9/printf/



Instead of re-implementing these specifiers in different ways and
because this method is not being used anywhere, it was completely
removed.

Signed-off-by: Or Cohen <orc@fewbytes.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

When building with -O0 we get:

 bsd/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_ioctl.o: In
 function `zfs_ioc_destroy_snaps_nvl':
 (...)/bsd/sys/cddl/contrib/opensolaris/uts/common/fs/zfs/zfs_ioctl.c:3265:
 undefined reference to `zvol_remove_minor'

Looks like the problem is that the missing symbol,
`zvol_remove_minor', referenced from zfs_ioctl.o is not in the zvol.o
because it's not defined, because it is inside '#ifdef NOTYET' block
(which is off). It works in the normal build because when compiled
with -O2 (and even with -O1), the reference to zvol_remove_minor in
zfs_ioctl.c is optimized away.

This patch puts the block using zvol_remove_minor also inside '#ifdef
NOTYET' to be consistent.

Signed-off-by: Tomasz Grabiec <tgrabiec@cloudius-systems.com>

FreeBSD networking code limits IP datagram size by IP_MAXPACKET
which is exactly 64K. This calculation doesn't take into account
Ethernet header length and may generate Ethernet packet longer than 64K.

Such a packet cannot be processed properly by some devices (Xen netback)
and being dropped.

This patch fixes this corner case for all supported IP protocols.

Signed-off-by: Dmitry Fleytman <dmitry@daynix.com>

This patch fixes part of TCP instability/low throughput problems on Xen.

On transmit networking stack passes to Xen netfront driver a list of mbufs
of up to PAGE_SIZE (4096) bytes each.
In case the list consists of more than MAX_TX_REQ_FRAGS (18) fragments
driver tries to defragment it with m_defrag().

m_defrag() in turn tries to build list of buffers of up to MCLBYTES
(2048) bytes each. This leads to even longer chain and packet being dropped.
As a result around 1% of TCP segments are lost and intensive TCP
retransmissions and slowdowns occur.

This patch makes m_defrag() use Jumbo packets zone for allocation, i.e.
PAGE_SIZE per mbuf.

This patch is pretty similar to what Glauber done on RX path before.

Signed-off-by: Dmitry Fleytman <dmitry@daynix.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

root mounts accept both a device and a pool name, while non-root mounts
only accept the pool name (the device(s) already known while mounting the pool
during the root mount).

Adjust the mount protocol for this.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

A zpool has a zfs as its userspace interface; but the call to create it was
disabled as part of the porting effort.

Re-enable it so osv can create zpools.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

MAXPATHLEN on osv is larger than on bsd, leading to the zfs ioctl command
buffer overflowing the maximum allowable size.

Reduce the buffer size to avoid a compile time assert failure.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

This is a shortcut through various kernel layers to avoid complication,
intended to be called from the various zfs utilities.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Not supported under osv.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Fixed MAXPATHLEN

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

We have _KERNEL defines scattered throughout the code, which makes
understanding it difficult.

Define it just once, and adjust the source to build.

We define it in an overridable variable, so that non-kernel imported code
can undo it.

Import FreeBSD files, changeset 245655.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Imported with no change from FreeBSD 245655.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

Signed-off-by: Benoit Canet <benoit@irqsave.net>
Signed-off-by: Avi Kivity <avi@cloudius-systems.com>

bsd's ifconf conflicts with osv's; rename it.

We use the bsd version in <osv/ioctl.h>, since we currently don't support
the Linux-ABI variants of these ioctls.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Tested-By: Benoit Canet <benoit@irqsave.net>

The bsd ifaddr struct conflicts with the osv ifaddr struct, which is a
public interface.  Rename the bsd struct to avoid conflict.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Tested-By: Benoit Canet <benoit@irqsave.net>

Workaround a bytorder function conflict, and reconcile a declaration.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Tested-By: Benoit Canet <benoit@irqsave.net>

Some structures are duplicated; move the duplicates to a common header
<netinet/__in.h>.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Tested-By: Benoit Canet <benoit@irqsave.net>

Some structures are duplicated; deduplicate them.

A few are source-compatible but not binary-compatible; use the ones from
<bits/socket.h>.

Others are both source- and binary- compatible; put them in a new header
<sys/__socket.h> which is included from both.

Work around a problem with the byteorder functions/macros.

Signed-off-by: Avi Kivity <avi@cloudius-systems.com>
Tested-By: Benoit Canet <benoit@irqsave.net>

Update ->va_nlink() in zfs_getattr() in preparation for sys_link().

Signed-off-by: Raphael S. Carvalho <raphael.scarv@gmail.com>
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

Wire up the VOP_LINK vnode operation for ZFS in preparation for
sys_link().

Signed-off-by: Raphael S. Carvalho <raphael.scarv@gmail.com>
[ penberg: drop FIGNORE, cleanup, split ]
Signed-off-by: Pekka Enberg <penberg@cloudius-systems.com>

* %lu => %u when reading flush. The barrier flag had the same bug, but I
ended up recreating it for flush.
* Move check of xb_flags to after we check sc->flags != NULL, as spotted by
Dima

Not all Xen versions implement the barrier feature in their pv disks. Such is
the case in Amazon EC2. For those, we should interpret the flush request and
wait for the requests until they are all handled. Also, we can still try to use
one of either barriers or flush-disk operations before we resort to guest-side
software implementation. Our driver currently only tests for one of them, so
this patch also implements flush requests.

Luckily, the implementation of xb_dump (which we don't currently use) needs to
do that as well, and this is also quite well isolated in xb_quiesce(). All we need
to do is call xb_quiesce() if flushing is not available in our backend

This patch implements GSI interrupt support for Xen bus.
Needed in Xen environments w/o vector callbacks for HVM.
One example of such an environment is Amazon EC2.

We cannot read the partition table from the device if the device is not marked
as ready, since all IO will stall. I believe it should be fine to just mark the
device ready before we mark our state as connected. With that change, it all
proceed normally.