mechanism so that early processing on mbufs can be performed before
a context switch to the NFS server threads. Because of this, if
the socket code is running without Giant, the NFS server also needs
to be able to run the upcall code without relying on the presence on
Giant. This change modifies the NFS server to run using a "giant
code lock" covering operation of the whole subsystem. Work is in
progress to move to data-based locking as part of the NFSv4 server
changes.
Introduce an NFS server subsystem lock, 'nfsd_mtx', and a set of
macros to operate on the lock:
NFSD_LOCK_ASSERT() Assert nfsd_mtx owned by current thread
NFSD_UNLOCK_ASSERT() Assert nfsd_mtx not owned by current thread
NFSD_LOCK_DONTCARE() Advisory: this function doesn't care
NFSD_LOCK() Lock nfsd_mtx
NFSD_UNLOCK() Unlock nfsd_mtx
Constify a number of global variables/structures in the NFS server
code, as they are not modified and contain constants only:
nfsrvv2_procid nfsrv_nfsv3_procid nonidempotent
nfsv2_repstat nfsv2_type nfsrv_nfsv3_procid
nfsrvv2_procid nfsrv_v2errmap nfsv3err_null
nfsv3err_getattr nfsv3err_setattr nfsv3err_lookup
nfsv3err_access nfsv3err_readlink nfsv3err_read
nfsv3err_write nfsv3err_create nfsv3err_mkdir
nfsv3err_symlink nfsv3err_mknod nfsv3err_remove
nfsv3err_rmdir nfsv3err_rename nfsv3err_link
nfsv3err_readdir nfsv3err_readdirplus nfsv3err_fsstat
nfsv3err_fsinfo nfsv3err_pathconf nfsv3err_commit
nfsrv_v3errmap
There are additional structures that should be constified but due
to their being passed into general purpose functions without const
arguments, I have not yet converted.
In general, acquire nfsd_mtx when accessing any of the global NFS
structures, including struct nfssvc_sock, struct nfsd, struct
nfsrv_descript.
Release nfsd_mtx whenever calling into VFS, and acquire Giant for
calls into VFS. Giant is not required for any part of the
operation of the NFS server with the exception of calls into VFS.
Giant will never by acquired in the upcall code path. However, it
may operate entirely covered by Giant, or not. If debug.mpsafenet
is set to 0, the system calls will acquire Giant across all
operations, and the upcall will assert Giant. As such, by default,
this enables locking and allows us to test assertions, but should not
cause any substantial new amount of code to be run without Giant.
Bugs should manifest in the form of lock assertion failures for now.
This approach is similar (but not identical) to modifications to the
BSD/OS NFS server code snapshot provided by BSDi as part of their
SMPng snapshot. The strategy is almost the same (single lock over
the NFS server), but differs in the following ways:
- Our NFS client and server code bases don't overlap, which means
both fewer bugs and easier locking (thanks Peter!). Also means
NFSD_*() as opposed to NFS_*().
- We make broad use of assertions, whereas the BSD/OS code does not.
- Made slightly different choices about how to handle macros building
packets but operating with side effects.
- We acquire Giant only when entering VFS from the NFS server daemon
threads.
- Serious bugs in BSD/OS implementation corrected -- the snapshot we
received was clearly a work in progress.
Based on ideas from: BSDi SMPng Snapshot
Reviewed by: rick@snowhite.cis.uoguelph.ca
Extensive testing by: kris
a ucred by itself as part of an nfs descriptor, then bzero's the ucred,
fails to initialize the mutex, etc. This is very bad, but I don't have
time to fix it right now. nfsd should instead hold a cred pointer,
and the credential should be properly initialized, probably from a
descendent of a kernel process credential.
system lockups (infinite loops) when a zero-length RPC is received.
Linux clients will sometimes send zero-length RPC requests.
Reorganize the use of recm in the loop.
Cc: security@freebsd.org
Submitted by: Mike Junk <junk@isilon.com>
MFC after: 3 days
o Add a mutex (sb_mtx) to struct sockbuf. This protects the data in a
socket buffer. The mutex in the receive buffer also protects the data
in struct socket.
o Determine the lock strategy for each members in struct socket.
o Lock down the following members:
- so_count
- so_options
- so_linger
- so_state
o Remove *_locked() socket APIs. Make the following socket APIs
touching the members above now require a locked socket:
- sodisconnect()
- soisconnected()
- soisconnecting()
- soisdisconnected()
- soisdisconnecting()
- sofree()
- soref()
- sorele()
- sorwakeup()
- sotryfree()
- sowakeup()
- sowwakeup()
Reviewed by: alfred
next to equivalent m_len adjustments. Move the nfsm_subs.h macros
into groups depending on which phase they are used in, since that
affects the error recovery requirements. Collect some of the common error
checking into a single macro as preparation for unwinding some more.
Have nfs_rephead return a value instead of secretly modifying args.
Remove some unused function arguments that were being passed around.
Clarify nfsm_reply()'s error handling (I hope).
Note ALL MODULES MUST BE RECOMPILED
make the kernel aware that there are smaller units of scheduling than the
process. (but only allow one thread per process at this time).
This is functionally equivalent to teh previousl -current except
that there is a thread associated with each process.
Sorry john! (your next MFC will be a doosie!)
Reviewed by: peter@freebsd.org, dillon@freebsd.org
X-MFC after: ha ha ha ha
other "system" header files.
Also help the deprecation of lockmgr.h by making it a sub-include of
sys/lock.h and removing sys/lockmgr.h form kernel .c files.
Sort sys/*.h includes where possible in affected files.
OK'ed by: bde (with reservations)
This is because calls with M_WAIT (now M_TRYWAIT) may not wait
forever when nothing is available for allocation, and may end up
returning NULL. Hopefully we now communicate more of the right thing
to developers and make it very clear that it's necessary to check whether
calls with M_(TRY)WAIT also resulted in a failed allocation.
M_TRYWAIT basically means "try harder, block if necessary, but don't
necessarily wait forever." The time spent blocking is tunable with
the kern.ipc.mbuf_wait sysctl.
M_WAIT is now deprecated but still defined for the next little while.
* Fix a typo in a comment in mbuf.h
* Fix some code that was actually passing the mbuf subsystem's M_WAIT to
malloc(). Made it pass M_WAITOK instead. If we were ever to redefine the
value of the M_WAIT flag, this could have became a big problem.
reserve, in maximal NFS packets. Originally only 2 packets worth of
space was reserved. The default is now 4, which appears to greatly
improve performance for slow to mid-speed machines on gigabit networks.
Add documentation and correct some prior documentation.
Problem Researched by: Andrew Gallatin <gallatin@cs.duke.edu>
Approved by: jkh
of element [4] in both, which goes beyond the end of the array, leaving
[0], [1], [2], and [3]. This bug did not cause any problems since
the overrun fields are initialized after the bogus array init but
needs to be fixed anyway.
Submitted by: Ian Dowse <iedowse@maths.tcd.ie>
-----------------------------
The core of the signalling code has been rewritten to operate
on the new sigset_t. No methodological changes have been made.
Most references to a sigset_t object are through macros (see
signalvar.h) to create a level of abstraction and to provide
a basis for further improvements.
The NSIG constant has not been changed to reflect the maximum
number of signals possible. The reason is that it breaks
programs (especially shells) which assume that all signals
have a non-null name in sys_signame. See src/bin/sh/trap.c
for an example. Instead _SIG_MAXSIG has been introduced to
hold the maximum signal possible with the new sigset_t.
struct sigprop has been moved from signalvar.h to kern_sig.c
because a) it is only used there, and b) access must be done
though function sigprop(). The latter because the table doesn't
holds properties for all signals, but only for the first NSIG
signals.
signal.h has been reorganized to make reading easier and to
add the new and/or modified structures. The "old" structures
are moved to signalvar.h to prevent namespace polution.
Especially the coda filesystem suffers from the change, because
it contained lines like (p->p_sigmask == SIGIO), which is easy
to do for integral types, but not for compound types.
NOTE: kdump (and port linux_kdump) must be recompiled.
Thanks to Garrett Wollman and Daniel Eischen for pressing the
importance of changing sigreturn as well.
txdr_hyper and fxdr_hyper tweaks to avoid excessive CPU order knowledge.
nfs_serv.c: don't call nfsm_adj() with negative values, windows clients
could crash servers when doing a readdir of a large directory.
nfs_socket.c: Use IP_PORTRANGE to get a priviliged port without a spin
loop trying to bind(). Don't clobber a mbuf pointer or we get panics
on a NFS3ERR_JUKEBOX error from a server when reusing a freed mbuf.
nfs_subs.c: Don't loose st_blocks on NFSv2 mounts when > 2GB.
Obtained from: OpenBSD
piecemeal, middle-of-file writes for NFS. These hacks have caused no
end of trouble, especially when combined with mmap(). I've removed
them. Instead, NFS will issue a read-before-write to fully
instantiate the struct buf containing the write. NFS does, however,
optimize piecemeal appends to files. For most common file operations,
you will not notice the difference. The sole remaining fragment in
the VFS/BIO system is b_dirtyoff/end, which NFS uses to avoid cache
coherency issues with read-merge-write style operations. NFS also
optimizes the write-covers-entire-buffer case by avoiding the
read-before-write. There is quite a bit of room for further
optimization in these areas.
The VM system marks pages fully-valid (AKA vm_page_t->valid =
VM_PAGE_BITS_ALL) in several places, most noteably in vm_fault. This
is not correct operation. The vm_pager_get_pages() code is now
responsible for marking VM pages all-valid. A number of VM helper
routines have been added to aid in zeroing-out the invalid portions of
a VM page prior to the page being marked all-valid. This operation is
necessary to properly support mmap(). The zeroing occurs most often
when dealing with file-EOF situations. Several bugs have been fixed
in the NFS subsystem, including bits handling file and directory EOF
situations and buf->b_flags consistancy issues relating to clearing
B_ERROR & B_INVAL, and handling B_DONE.
getblk() and allocbuf() have been rewritten. B_CACHE operation is now
formally defined in comments and more straightforward in
implementation. B_CACHE for VMIO buffers is based on the validity of
the backing store. B_CACHE for non-VMIO buffers is based simply on
whether the buffer is B_INVAL or not (B_CACHE set if B_INVAL clear,
and vise-versa). biodone() is now responsible for setting B_CACHE
when a successful read completes. B_CACHE is also set when a bdwrite()
is initiated and when a bwrite() is initiated. VFS VOP_BWRITE
routines (there are only two - nfs_bwrite() and bwrite()) are now
expected to set B_CACHE. This means that bowrite() and bawrite() also
set B_CACHE indirectly.
There are a number of places in the code which were previously using
buf->b_bufsize (which is DEV_BSIZE aligned) when they should have
been using buf->b_bcount. These have been fixed. getblk() now clears
B_DONE on return because the rest of the system is so bad about
dealing with B_DONE.
Major fixes to NFS/TCP have been made. A server-side bug could cause
requests to be lost by the server due to nfs_realign() overwriting
other rpc's in the same TCP mbuf chain. The server's kernel must be
recompiled to get the benefit of the fixes.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
routine was [ab]used for two different things, and you couldn't tell from
the wait channel which one had wedged.
Catch a few things missing from NFS_NOSERVER.
is less than NFS_MINPACKET or greater than NFS_MAXPACKET in size, it
barfs and, I think, drops the connection.
However, there's no guarantee that in a multi-fragment RPC, all the
fragments will be at least as large as NFS_MINPACKET.
In fact, with the version of "tclnfs" we have here, which supports NFS
over TCP, at least when built under SunOS 4.1.3 (i.e., with 4.1.3's
user-mode ONC RPC library), I can *repeatably* cause "tclnfs" to send a
request with more than one fragment, one of which is only 8 bytes long.
I just do a 3877-byte write to a file, at an offset of 0.
The check that "slp->ns_reclen" is greater than or equal to
NFS_MINPACKET serves no useful purpose - if the NFS server code can't
handle packets < NFS_MINPACKET bytes, it can't handle them over *any*
protocol, so the check has to be done above the RPC-over-TCP layer - and
should be removed.
Obtained from: Fix from Guy Harris, forwarded by Rick Macklem.
detachment of vfs sysctls. Unloading of vfs LKMs doesn't actually
work for any vfs, since it leaves garbage pointers to memory
allocation control structures.
