-Fix an mbuf leak caused by a cut&paste bug where the small ring's mbufs
were never freed, but the big ring was freed twice.
-Don't supply rx hw csums for frames which are padded beyond the
length specified in the ip header. If the padding is non-zero,
the hw csum will be incorrect for such frames.
Increase the verbosity of the warning given when this file is used as
part of the old tty system - helping the user to know how to transition to
the new tty api.
MFp4: Optimize mappedwrite() and mappedread() functions to write/read as much
non-mapped data as possible at once and not page-by-page. Which this change we
combain I/Os, but also saves many VM_OBJECT_UNLOCK()/VM_OBJECT_LOCK()
operations.
Use the same timeout parameters for BIO_FLUSH as with the other commands.
This workaround the problem in Parallels/VMWare where the emulated drivers are
slower, especially with ATA_FLUSHCACHE. The problem appears much more
frequently with ZFS which use it a lot more.
- Always try to write one whole page at a time.
- vm_page_undirty() is enough (instead of vm_page_set_validclean()), but it has
to be called before we write the data in case someone makes page dirty after
our write, but before our vm_page_undirty() call.
- Always dmu_write, not matter if uiomove() succeeded, because it could
partially be ok and we would lose some changes.
Allow the dounmount() to proceed even for doomed coveredvp.
In dounmount(), before or while vn_lock(coveredvp) is called, coveredvp
vnode may be VI_DOOMED due to one of the following:
- other thread finished unmount and vput()ed it, and vnode was chosen
for recycling, while vn_lock() slept;
- forced unmount of the coveredvp->v_mount fs.
In the first case, next check for changed v_mountedhere or mnt_gen counter
would be successfull. In the second case, the unmount shall be allowed.
Various fixes to the NFS Directio support.
- Fix for a bug where a close would not wait for all (directio)
dirty buffers to drain. The nfsnode was not marked NMODIFIED
when there were directio dirtied buffers pending, causing this.
- No reason to vhold/vrele the vp when enqueueing DirectIO requests
for the nfsiods. The vnode can't really go way since the close
has to wait for these requests to drain.
Invalidate all TLBs and page structure caches that may reference a page
for page walk acceleration before releasing it to the free list.
This is required for CPUs implementing page structure caches as described
in the Intel application note:
"TLBs, Paging-Structure Caches, and Their Invalidation"
(Document Number: 317080-001)
Use a tighter check to see if a resource allocation request is for a
specific request and thus should first try to be allocated from the
sys_resource pool. This avoids using the sys_resource pool for wildcard
requests that have bounded ranges coming from cbb(4) and Host-PCI pcib(4)
drivers.
Tested by: Andrea Bittau <a.bittau of cs.ucl.ac.uk fame>
Sleuthing by: Andrea Bittau as well
- HT 2.00b added a new flag to the MSI mapping HT capability to indicate
that the MSI mapping window is fixed at 0xfee00000 and the capability
does not include two more dwords used to program the address. Supporting
this mostly results in quieting spurious warnings during boot about
non-default MSI mapping windows.
- HT 2.00b also added a new HT capability type, so support that in pciconf.
Work around GMAC hardware hang bug.
It seems that valid pause frames(Tx flow control) cause GMAC to hang
such that it resulted in watchdog timeout. As a work around don't
flush Rx MAC FIFO if we've received pause frames.
Tested by: Harald Schmalzbauer (h DOT schmalzbauer AT omnisec DOT de)
Disable TSO support.
Under certain circumtances, if TSO is active, Yukon II generates
corrupted IP packets. All corrupted IP packets I noticed were the the
last segmented packet in a TSO request. The corrupted packet resulted
in retransmission of the damaged packet which in turn decreased network
performance dramatically.
Unfortunately it seems that there is no way to workaround this bug
as TSO is completely handled in hardware. Disable TSO until we find a
working workaround or a new silicon revision that doesn't have this
hardware bug.
Fix the triple fault used as a last resort during a reboot to actually
fault. The previous method zero'd out the page tables, invalidated the
TLB, and then entered a spin loop. The idea was that the instruction after
the TLB invalidate would result in a page fault and the page fault and
subsequent double fault wouldn't be able to determine the physical page
for their fault handlers' first instruction. This stopped working when
PGE (PG_G PTE/PDE bit) support was added as a TLB invalidate via %cr3
reload doesn't clear TLB entries with PG_G set. Thus, the CPU was still
able to map the virtual address for the spin loop and happily performed
its infinite loop.
The triple fault now uses a much more deterministic sledge-hammer approach
to generate a triple fault. First, the IDT descriptor is set to point to
an empty IDT, so any interrupts (including a double fault) will instantly
fault. Second, we trigger a int 3 breakpoint to force an interrupt and
kick off a triple fault.
MFi386: Attempt to reset the machine using the Reset Control register and
Fast A20 and Init register if the keyboard reset doesn't work before
resorting to a triple fault.
ZIL (ZFS Intent Log) can be safely turned on and off at run time, because
it is only used when dataset is beeing mounted to decide if log should also
be opened.
MFp4: Rearange the code so vobject is destroyed from reclaim() method like
in all other file system on FreeBSD (instead from inactive() method).
A nice side-effect of this change, except that it speedups file system
when mmaped file are often open/closed, is that it makes FreeBSD's
namecache work:)
MFp4: Once page is written successfully, we should clear the dirty bits.
This fixes slow operations on mmaped files, because without this fix,
pages were written to disk multiple times.
If one is looking for even greater speed up for such operation, he should
disable ZIL (by setting vfs.zfs.zil_disable to 1 in /boot/loader.conf).
Disabling ZIL makes fsx run ~9 times faster.
Avoid false assertion on transmit and delayed ack timeout with enabled invariants.
Replace callout_pending() by callout_active() to remove race window.
Rename mac*devfsdirent*() to mac*devfs*() to synchronize with SEDarwin,
where similar data structures exist to support devfs and the MAC
Framework, but are named differently.
Obtained from: TrustedBSD Project
Sponsored by: SPARTA, Inc.
Turn off route header processing for now due to issues pointed out
by Philippe Biondi and Arnaud Ebalard. This is a temporary fix
until more discussion can be had on the exact risks involved in
allowing source routing in IPv6
Submitted by: itojun
Reviewed by: jinmei
MFC after: 1 day
sam [Mon, 23 Apr 2007 05:51:18 +0000 (05:51 +0000)]
o fix a buffer overflow in save_key() that occurs with 104-bit wep
o make some variables parameters (frequency of crack, and maximum channel)
o try to spoof mac if association fails
Submitted by: Andrea Bittau <a.bittau@cs.ucl.ac.uk>
MFp4: Reduce diff against vendor code:
- Move FreeBSD-specific code to zfs_freebsd_*() functions in zfs_vnops.c
and keep original functions as similar to vendor's code as possible.
- Add various includes back, now that we have them.
Add back the original behavior of changing the entire directory path at
once (CWD a/b/c vs. 3 CWDs). If an error occurs, we fall back to the default
method of a single CWD per directory element. Since this is technically
a violation of the basic FTP RFC, this behavior is under a compile-time
option FTP_COMBINE_CWDS and is off by default. It should work with most
Unix-based FTP daemons and can save latency.
Fix 'zpool status -v'. To get object number we should use ZFS_DIRENT_OBJ()
macro, as za_first_integer field also contains type. This should be fixed in
ZFS itself, but this bug is not visible on Solaris, because there, type is
not stored in za_first_integer. On the other hand it will be visible on
MacOS X.
When zfs dataset has jailed=on property, it won't be mounted with
'zfs mount -a' from the main system - this is by design, as mountpoint
may be set to dangerous value. This all means, that such file system
has to be mounted from within a jail. To make it easier, reorganize
rc.d/zfs script so it can be used from within a jail.
Normalize variable naming in the MAC Framework by adopting the normal
variable name conventions for arguments passed into the framework --
for example, name network interfaces 'ifp', sockets 'so', mounts 'mp',
mbufs 'm', processes 'p', etc, wherever possible. Previously there
was significant variation in this regard.
Normalize copyright lists to ranges where sensible.
In the MAC Framework implementation, file systems have two per-mountpoint
labels: the mount label (label of the mountpoint) and the fs label (label
of the file system). In practice, policies appear to only ever use one,
and the distinction is not helpful.
Combine mnt_mntlabel and mnt_fslabel into a single mnt_label, and
eliminate extra machinery required to maintain the additional label.
Update policies to reflect removal of extra entry points and label.
Obtained from: TrustedBSD Project
Sponsored by: SPARTA, Inc.
Remove MAC Framework access control check entry points made redundant with
the introduction of priv(9) and MAC Framework entry points for privilege
checking/granting. These entry points exactly aligned with privileges and
provided no additional security context:
Add mpo_priv_check() implementations to Biba and LOMAC policies, which,
for each privilege, determine if they can be granted to processes
considered unprivileged by those two policies. These mostly, but not
entirely, align with the set of privileges granted in jails.