it will work with either the old or new server.
The FHA code keeps a cache of currently active file handles for
NFSv2 and v3 requests, so that read and write requests for the same
file are directed to the same group of threads (reads) or thread
(writes). It does not currently work for NFSv4 requests. They are
more complex, and will take more work to support.
This improves read-ahead performance, especially with ZFS, if the
FHA tuning parameters are configured appropriately. Without the
FHA code, concurrent reads that are part of a sequential read from
a file will be directed to separate NFS threads. This has the
effect of confusing the ZFS zfetch (prefetch) code and makes
sequential reads significantly slower with clients like Linux that
do a lot of prefetching.
The FHA code has also been updated to direct write requests to nearby
file offsets to the same thread in the same way it batches reads,
and the FHA code will now also send writes to multiple threads when
needed.
This improves sequential write performance in ZFS, because writes
to a file are now more ordered. Since NFS writes (generally
less than 64K) are smaller than the typical ZFS record size
(usually 128K), out of order NFS writes to the same block can
trigger a read in ZFS. Sending them down the same thread increases
the odds of their being in order.
In order for multiple write threads per file in the FHA code to be
useful, writes in the NFS server have been changed to use a LK_SHARED
vnode lock, and upgrade that to LK_EXCLUSIVE if the filesystem
doesn't allow multiple writers to a file at once. ZFS is currently
the only filesystem that allows multiple writers to a file, because
it has internal file range locking. This change does not affect the
NFSv4 code.
This improves random write performance to a single file in ZFS, since
we can now have multiple writers inside ZFS at one time.
I have changed the default tuning parameters to a 22 bit (4MB)
window size (from 256K) and unlimited commands per thread as a
result of my benchmarking with ZFS.
The FHA code has been updated to allow configuring the tuning
parameters from loader tunable variables in addition to sysctl
variables. The read offset window calculation has been slightly
modified as well. Instead of having separate bins, each file
handle has a rolling window of bin_shift size. This minimizes
glitches in throughput when shifting from one bin to another.
sys/conf/files:
Add nfs_fha_new.c and nfs_fha_old.c. Compile nfs_fha.c
when either the old or the new NFS server is built.
sys/fs/nfs/nfsport.h,
sys/fs/nfs/nfs_commonport.c:
Bring in changes from Rick Macklem to newnfs_realign that
allow it to operate in blocking (M_WAITOK) or non-blocking
(M_NOWAIT) mode.
sys/fs/nfs/nfs_commonsubs.c,
sys/fs/nfs/nfs_var.h:
Bring in a change from Rick Macklem to allow telling
nfsm_dissect() whether or not to wait for mallocs.
sys/fs/nfs/nfsm_subs.h:
Bring in changes from Rick Macklem to create a new
nfsm_dissect_nonblock() inline function and
NFSM_DISSECT_NONBLOCK() macro.
sys/fs/nfs/nfs_commonkrpc.c,
sys/fs/nfsclient/nfs_clkrpc.c:
Add the malloc wait flag to a newnfs_realign() call.
sys/fs/nfsserver/nfs_nfsdkrpc.c:
Setup the new NFS server's RPC thread pool so that it will
call the FHA code.
Add the malloc flag argument to newnfs_realign().
Unstaticize newnfs_nfsv3_procid[] so that we can use it in
the FHA code.
sys/fs/nfsserver/nfs_nfsdsocket.c:
In nfsrvd_dorpc(), add NFSPROC_WRITE to the list of RPC types
that use the LK_SHARED lock type.
sys/fs/nfsserver/nfs_nfsdport.c:
In nfsd_fhtovp(), if we're starting a write, check to see
whether the underlying filesystem supports shared writes.
If not, upgrade the lock type from LK_SHARED to LK_EXCLUSIVE.
sys/nfsserver/nfs_fha.c:
Remove all code that is specific to the NFS server
implementation. Anything that is server-specific is now
accessed through a callback supplied by that server's FHA
shim in the new softc.
There are now separate sysctls and tunables for the FHA
implementations for the old and new NFS servers. The new
NFS server has its tunables under vfs.nfsd.fha, the old
NFS server's tunables are under vfs.nfsrv.fha as before.
In fha_extract_info(), use callouts for all server-specific
code. Getting file handles and offsets is now done in the
individual server's shim module.
In fha_hash_entry_choose_thread(), change the way we decide
whether two reads are in proximity to each other.
Previously, the calculation was a simple shift operation to
see whether the offsets were in the same power of 2 bucket.
The issue was that there would be a bucket (and therefore
thread) transition, even if the reads were in close
proximity. When there is a thread transition, reads wind
up going somewhat out of order, and ZFS gets confused.
The new calculation simply tries to see whether the offsets
are within 1 << bin_shift of each other. If they are, the
reads will be sent to the same thread.
The effect of this change is that for sequential reads, if
the client doesn't exceed the max_reqs_per_nfsd parameter
and the bin_shift is set to a reasonable value (22, or
4MB works well in my tests), the reads in any sequential
stream will largely be confined to a single thread.
Change fha_assign() so that it takes a softc argument. It
is now called from the individual server's shim code, which
will pass in the softc.
Change fhe_stats_sysctl() so that it takes a softc
parameter. It is now called from the individual server's
shim code. Add the current offset to the list of things
printed out about each active thread.
Change the num_reads and num_writes counters in the
fha_hash_entry structure to 32-bit values, and rename them
num_rw and num_exclusive, respectively, to reflect their
changed usage.
Add an enable sysctl and tunable that allows the user to
disable the FHA code (when vfs.XXX.fha.enable = 0). This
is useful for before/after performance comparisons.
nfs_fha.h:
Move most structure definitions out of nfs_fha.c and into
the header file, so that the individual server shims can
see them.
Change the default bin_shift to 22 (4MB) instead of 18
(256K). Allow unlimited commands per thread.
sys/nfsserver/nfs_fha_old.c,
sys/nfsserver/nfs_fha_old.h,
sys/fs/nfsserver/nfs_fha_new.c,
sys/fs/nfsserver/nfs_fha_new.h:
Add shims for the old and new NFS servers to interface with
the FHA code, and callbacks for the
The shims contain all of the code and definitions that are
specific to the NFS servers.
They setup the server-specific callbacks and set the server
name for the sysctl and loader tunable variables.
sys/nfsserver/nfs_srvkrpc.c:
Configure the RPC code to call fhaold_assign() instead of
fha_assign().
sys/modules/nfsd/Makefile:
Add nfs_fha.c and nfs_fha_new.c.
sys/modules/nfsserver/Makefile:
Add nfs_fha_old.c.
Reviewed by: rmacklem
Sponsored by: Spectra Logic
MFC after: 2 weeks
- Don't insert BKGRDMARKER bufs into the splay or dirty/clean buf lists.
No consumers need to find them there and it complicates the tree.
These flags are all FFS specific and could be moved out of the buf
cache.
- Use pbgetvp() and pbrelvp() to associate the background and journal
bufs with the vp. Not only is this much cheaper it makes more sense
for these transient bufs.
- Fix the assertions in pbget* and pbrel*. It's not safe to check list
pointers which were never initialized. Use the BX flags instead. We
also check B_PAGING in reassignbuf() so this should cover all cases.
Discussed with: kib, mckusick, attilio
Sponsored by: EMC / Isilon Storage Division
u_long. Before this change it was of type int for syscalls, but prototypes
in sys/stat.h and documentation for chflags(2) and fchflags(2) (but not
for lchflags(2)) stated that it was u_long. Now some related functions
use u_long type for flags (strtofflags(3), fflagstostr(3)).
- Make path argument of type 'const char *' for consistency.
Discussed on: arch
Sponsored by: The FreeBSD Foundation
locked. vnode_pager_setsize() might sleep waiting for the page after
EOF be unbusied.
Call vnode_pager_setsize() both for the regular and directory vnodes.
Reported by: mich
Reviewed by: rmacklem
Discussed with: avg, jhb
MFC after: 2 weeks
In common configurations biosize is a power of two, but is not required to
be so. Thanks to markj@ for spotting an additional case beyond my original
patch.
Reviewed by: rmacklem@
cluster_write() and cluster_wbuild() functions. The flags to be
allowed are a subset of the GB_* flags for getblk().
Sponsored by: The FreeBSD Foundation
Tested by: pho
proper locking. This doesn't prevent in any case reclaim of the vnode.
Avoid this not going over-the-wire in this case and relying on subsequent
smbfs_getattr() call to restore consistency.
While I'm here, change a couple of SMBVDEBUG() in MPASS().
sbmfs_smb_lookup() doesn't and shouldn't know about '.' and '..'
Reported by: pho's stress2 suite
future further optimizations where the vm_object lock will be held
in read mode most of the time the page cache resident pool of pages
are accessed for reading purposes.
The change is mostly mechanical but few notes are reported:
* The KPI changes as follow:
- VM_OBJECT_LOCK() -> VM_OBJECT_WLOCK()
- VM_OBJECT_TRYLOCK() -> VM_OBJECT_TRYWLOCK()
- VM_OBJECT_UNLOCK() -> VM_OBJECT_WUNLOCK()
- VM_OBJECT_LOCK_ASSERT(MA_OWNED) -> VM_OBJECT_ASSERT_WLOCKED()
(in order to avoid visibility of implementation details)
- The read-mode operations are added:
VM_OBJECT_RLOCK(), VM_OBJECT_TRYRLOCK(), VM_OBJECT_RUNLOCK(),
VM_OBJECT_ASSERT_RLOCKED(), VM_OBJECT_ASSERT_LOCKED()
* The vm/vm_pager.h namespace pollution avoidance (forcing requiring
sys/mutex.h in consumers directly to cater its inlining functions
using VM_OBJECT_LOCK()) imposes that all the vm/vm_pager.h
consumers now must include also sys/rwlock.h.
* zfs requires a quite convoluted fix to include FreeBSD rwlocks into
the compat layer because the name clash between FreeBSD and solaris
versions must be avoided.
At this purpose zfs redefines the vm_object locking functions
directly, isolating the FreeBSD components in specific compat stubs.
The KPI results heavilly broken by this commit. Thirdy part ports must
be updated accordingly (I can think off-hand of VirtualBox, for example).
Sponsored by: EMC / Isilon storage division
Reviewed by: jeff
Reviewed by: pjd (ZFS specific review)
Discussed with: alc
Tested by: pho
from the tree since few months (please note that the userland bits
were already disconnected since a long time, thus there is no need
to update the OLD* entries).
This is not targeted for MFC.
Include some flags of the nullfs mount itself:
MNT_RDONLY, MNT_NOEXEC, MNT_NOSUID, MNT_UNION, MNT_NOSYMFOLLOW.
This allows userland code calling statfs() or fstatfs() to see these flags.
In particular, this allows opendir() to detect that a -t nullfs -o union
mount needs deduplication (otherwise at least . and .. are returned twice)
and allows rtld to detect a -t nullfs -o noexec mount as noexec.
Turn off the MNT_ROOTFS flag from the underlying filesystem because the
nullfs mount is definitely not the root filesystem.
Reviewed by: kib
MFC after: 1 week
- Capability is no longer separate descriptor type. Now every descriptor
has set of its own capability rights.
- The cap_new(2) system call is left, but it is no longer documented and
should not be used in new code.
- The new syscall cap_rights_limit(2) should be used instead of
cap_new(2), which limits capability rights of the given descriptor
without creating a new one.
- The cap_getrights(2) syscall is renamed to cap_rights_get(2).
- If CAP_IOCTL capability right is present we can further reduce allowed
ioctls list with the new cap_ioctls_limit(2) syscall. List of allowed
ioctls can be retrived with cap_ioctls_get(2) syscall.
- If CAP_FCNTL capability right is present we can further reduce fcntls
that can be used with the new cap_fcntls_limit(2) syscall and retrive
them with cap_fcntls_get(2).
- To support ioctl and fcntl white-listing the filedesc structure was
heavly modified.
- The audit subsystem, kdump and procstat tools were updated to
recognize new syscalls.
- Capability rights were revised and eventhough I tried hard to provide
backward API and ABI compatibility there are some incompatible changes
that are described in detail below:
CAP_CREATE old behaviour:
- Allow for openat(2)+O_CREAT.
- Allow for linkat(2).
- Allow for symlinkat(2).
CAP_CREATE new behaviour:
- Allow for openat(2)+O_CREAT.
Added CAP_LINKAT:
- Allow for linkat(2). ABI: Reuses CAP_RMDIR bit.
- Allow to be target for renameat(2).
Added CAP_SYMLINKAT:
- Allow for symlinkat(2).
Removed CAP_DELETE. Old behaviour:
- Allow for unlinkat(2) when removing non-directory object.
- Allow to be source for renameat(2).
Removed CAP_RMDIR. Old behaviour:
- Allow for unlinkat(2) when removing directory.
Added CAP_RENAMEAT:
- Required for source directory for the renameat(2) syscall.
Added CAP_UNLINKAT (effectively it replaces CAP_DELETE and CAP_RMDIR):
- Allow for unlinkat(2) on any object.
- Required if target of renameat(2) exists and will be removed by this
call.
Removed CAP_MAPEXEC.
CAP_MMAP old behaviour:
- Allow for mmap(2) with any combination of PROT_NONE, PROT_READ and
PROT_WRITE.
CAP_MMAP new behaviour:
- Allow for mmap(2)+PROT_NONE.
Added CAP_MMAP_R:
- Allow for mmap(PROT_READ).
Added CAP_MMAP_W:
- Allow for mmap(PROT_WRITE).
Added CAP_MMAP_X:
- Allow for mmap(PROT_EXEC).
Added CAP_MMAP_RW:
- Allow for mmap(PROT_READ | PROT_WRITE).
Added CAP_MMAP_RX:
- Allow for mmap(PROT_READ | PROT_EXEC).
Added CAP_MMAP_WX:
- Allow for mmap(PROT_WRITE | PROT_EXEC).
Added CAP_MMAP_RWX:
- Allow for mmap(PROT_READ | PROT_WRITE | PROT_EXEC).
Renamed CAP_MKDIR to CAP_MKDIRAT.
Renamed CAP_MKFIFO to CAP_MKFIFOAT.
Renamed CAP_MKNODE to CAP_MKNODEAT.
CAP_READ old behaviour:
- Allow pread(2).
- Disallow read(2), readv(2) (if there is no CAP_SEEK).
CAP_READ new behaviour:
- Allow read(2), readv(2).
- Disallow pread(2) (CAP_SEEK was also required).
CAP_WRITE old behaviour:
- Allow pwrite(2).
- Disallow write(2), writev(2) (if there is no CAP_SEEK).
CAP_WRITE new behaviour:
- Allow write(2), writev(2).
- Disallow pwrite(2) (CAP_SEEK was also required).
Added convinient defines:
#define CAP_PREAD (CAP_SEEK | CAP_READ)
#define CAP_PWRITE (CAP_SEEK | CAP_WRITE)
#define CAP_MMAP_R (CAP_MMAP | CAP_SEEK | CAP_READ)
#define CAP_MMAP_W (CAP_MMAP | CAP_SEEK | CAP_WRITE)
#define CAP_MMAP_X (CAP_MMAP | CAP_SEEK | 0x0000000000000008ULL)
#define CAP_MMAP_RW (CAP_MMAP_R | CAP_MMAP_W)
#define CAP_MMAP_RX (CAP_MMAP_R | CAP_MMAP_X)
#define CAP_MMAP_WX (CAP_MMAP_W | CAP_MMAP_X)
#define CAP_MMAP_RWX (CAP_MMAP_R | CAP_MMAP_W | CAP_MMAP_X)
#define CAP_RECV CAP_READ
#define CAP_SEND CAP_WRITE
#define CAP_SOCK_CLIENT \
(CAP_CONNECT | CAP_GETPEERNAME | CAP_GETSOCKNAME | CAP_GETSOCKOPT | \
CAP_PEELOFF | CAP_RECV | CAP_SEND | CAP_SETSOCKOPT | CAP_SHUTDOWN)
#define CAP_SOCK_SERVER \
(CAP_ACCEPT | CAP_BIND | CAP_GETPEERNAME | CAP_GETSOCKNAME | \
CAP_GETSOCKOPT | CAP_LISTEN | CAP_PEELOFF | CAP_RECV | CAP_SEND | \
CAP_SETSOCKOPT | CAP_SHUTDOWN)
Added defines for backward API compatibility:
#define CAP_MAPEXEC CAP_MMAP_X
#define CAP_DELETE CAP_UNLINKAT
#define CAP_MKDIR CAP_MKDIRAT
#define CAP_RMDIR CAP_UNLINKAT
#define CAP_MKFIFO CAP_MKFIFOAT
#define CAP_MKNOD CAP_MKNODAT
#define CAP_SOCK_ALL (CAP_SOCK_CLIENT | CAP_SOCK_SERVER)
Sponsored by: The FreeBSD Foundation
Reviewed by: Christoph Mallon <christoph.mallon@gmx.de>
Many aspects discussed with: rwatson, benl, jonathan
ABI compatibility discussed with: kib
changes in r246417 were incomplete as they did not add explicit calls to
sigdeferstop() around all the places that previously passed SBDRY to
_sleep(). In addition, nfs_getcacheblk() could trigger a write RPC from
getblk() resulting in sigdeferstop() recursing. Rather than manually
deferring stop signals in specific places, change the VFS_*() and VOP_*()
methods to defer stop signals for filesystems which request this behavior
via a new VFCF_SBDRY flag. Note that this has to be a VFC flag rather than
a MNTK flag so that it works properly with VFS_MOUNT() when the mount is
not yet fully constructed. For now, only the NFS clients are set this new
flag in VFS_SET().
A few other related changes:
- Add an assertion to ensure that TDF_SBDRY doesn't leak to userland.
- When a lookup request uses VOP_READLINK() to follow a symlink, mark
the request as being on behalf of the thread performing the lookup
(cnp_thread) rather than using a NULL thread pointer. This causes
NFS to properly handle signals during this VOP on an interruptible
mount.
PR: kern/176179
Reported by: Russell Cattelan (sigdeferstop() recursion)
Reviewed by: kib
MFC after: 1 month
is excessive. Postpone the flush of the fsinfo to VFS_SYNC(),
remembering the need for update with the flag MSDOSFS_FSIMOD, stored
in pm_flags.
FAT32 specification describes both FSI_Free_Count and FSI_Nxt_Free as
the advisory hints, not requiring them to be correct.
Based on the patch from bde, modified by me.
Reviewed by: bde
MFC after: 2 weeks
tmpfs_mapped{read, write}() functions:
- tmpfs_mapped{read, write}() are only called within VOP_{READ, WRITE}(),
which check before-hand to work only on valid VREG vnodes. Also the
vnode is locked for the duration of the work, making vnode reclaiming
impossible, during the operation. Hence, vobj can never be NULL.
- Currently check on resident pages and cached pages without vm object
lock held is racy and can do even more harm than good, as a page could
be transitioning between these 2 pools and then be skipped entirely.
Skip the checks as lookups on empty splay trees are very cheap.
Discussed with: alc
Tested by: flo
MFC after: 2 weeks
e2fs_maxcontig was modelled after UFS when bringing the
"Orlov allocator" to ext2. On UFS fs_maxcontig is kept in the
superblock and is used by userland tools (fsck and growfs),
In ext2 this information is volatile so it is not available
for userland tools, so in this case it doesn't have sense
to carry it in the in-memory superblock.
Also remove a pointless check for MAX(1, x) > 0.
Submitted by: Christoph Mallon
MFC after: 2 weeks
Posix requires that open(2) is restartable for SA_RESTART.
For non-posix objects, in particular, devfs nodes, still disable
automatic restart of the opens. The open call to a driver could have
significant side effects for the hardware.
Noted and reviewed by: jilles
Discussed with: bde
MFC after: 2 weeks
195702, 195703, and 195821 prevented a thread from suspending while holding
locks inside of NFS by forcing the thread to fail sleeps with EINTR or
ERESTART but defer the thread suspension to the user boundary. However,
this had the effect that stopping a process during an NFS request could
abort the request and trigger EINTR errors that were visible to userland
processes (previously the thread would have suspended and completed the
request once it was resumed).
This change instead effectively masks stop signals while in the NFS client.
It uses the existing TDF_SBDRY flag to effect this since SIGSTOP cannot
be masked directly. Also, instead of setting PBDRY on individual sleeps,
the NFS client now sets the TDF_SBDRY flag around each NFS request and
stop signals are masked for all sleeps during that region (the previous
change missed sleeps in lockmgr locks). The end result is that stop
signals sent to threads performing an NFS request are completely
ignored until after the NFS request has finished processing and the
thread prepares to return to userland. This restores the behavior of
stop signals being transparent to userland processes while still
preventing threads from suspending while holding NFS locks.
Reviewed by: kib
MFC after: 1 month
- Remove unused extern declarations in fs.h
- Correct comments in ext2_dir.h
- Several panic() messages showed wrong function names.
- Remove commented out stray line in ext2_alloc.c.
- Remove the unused macro EXT2_BLOCK_SIZE_BITS() and the then
write-only member e2fs_blocksize_bits from struct m_ext2fs.
- Remove the unused macro EXT2_FIRST_INO() and the then write-only
member e2fs_first_inode from struct m_ext2fs.
- Remove EXT2_DESC_PER_BLOCK() and the member e2fs_descpb from
struct m_ext2fs.
- Remove the unused members e2fs_bmask, e2fs_dbpg and
e2fs_mount_opt from struct m_ext2fs
- Correct harmless off-by-one error for fspath in ext2_vfsops.c.
- Remove the unused and broken macros EXT2_ADDR_PER_BLOCK_BITS()
and EXT2_DESC_PER_BLOCK_BITS().
- Remove the !_KERNEL versions of the EXT2_* macros.
Submitted by: Christoph Mallon
MFC after: 2 weeks
track the MNT_SYNCHRONOUS flag. It is set to the latter at mount time
but not updated by MNT_UPDATE.
Use MNT_SYNCHRONOUS to decide to write the FAT updates syncrhonously.
Submitted by: bde
MFC after: 1 week
a directory to a subdir of the root directory from somewhere else.
For all directory moves that change the parent directory, the dotdot
entry must be fixed up. For msdosfs, the root directory is magic for
non-FAT32. It is less magic for FAT32, but needs the same magic for
the dotdot fixup. It didn't have it.
Both chkdsk and fsck_msdosfs fix the corrupt directory entries with no
problems.
The fix is to use the same magic for dotdot in msdosfs_rename() as in
msdosfs_mkdir().
For msdosfs_mkdir(), document the magic. When writing the dotdot entry
in mkdir, use explicitly set pcl variable instead on relying on the
start cluster of the root directory typically has a value < 65536.
Submitted by: bde
MFC after: 1 week
Trying FAT32 on a small partition failed to mount because
pmp->pm_Sectors was nonzero. Normally, FAT32 file systems are so
large that the 16-bit pm_Sectors can't hold the size. This is
indicated by setting it to 0 and using only pm_HugeSectors. But at
least old versions of newfs_msdos use the 16-bit field if possible,
and msdosfs supports this except for breaking its own support in the
sanity check. This is quite different from the handling of pm_FATsecs
-- now the 16-bit value is always ignored for FAT32 except for
checking that it is 0, and newfs_msdos doesn't use the 16-bit value
for FAT32.
Submitted by: bde
MFC after: 1 week
this check is somewhere in the network code, but this assertion
already proven to be useful in catching what seems to be driver bugs
causing NFS scrambling random memory.
Discussed with: rmacklem
MFC after: 1 week
requested from the server for the read operation. Server shall not
reply with too large size, but client should be resilent too.
Reviewed by: rmacklem
MFC after: 1 week
- Use NFSD_MONOSEC (which maps to time_uptime) instead of the seconds
portion of wall-time stamps to manage timeouts on events.
- Remove unused nd_starttime from the per-request structure in the new
NFS server.
- Use nanotime() for the modification time on a delegation to get as
precise a time as possible.
- Use time_second instead of extracting the second from a call to
getmicrotime().
Submitted by: bde (3)
Reviewed by: bde, rmacklem
MFC after: 2 weeks
The previous change accidentally left the substraction we
were trying to avoid in case that i_blocks could become
negative.
Reported by: bde
MFC after: 4 days
Ext2fs uses unsigned fields in its dinode struct.
FreeBSD can have negative values in some of those
fields and the inode is meant to interact with the
system so we have never respected the unsigned
nature of most of those fields.
Block numbers and the NFS generation number do
not need to be signed so redefine them as
unsigned to better match the on-disk information.
MFC after: 1 week
Testing with fsx has revealed problems and in order to
hunt the bugs properly we need reduce the complexity.
This seems to help but is not a complete solution.
MFC after: 3 days
to the old one's nfs.nfsrv.async.
Please note that by enabling this option (default is disabled), the system
could potentionally have silent data corruption if the server crashes
before write is committed to non-volatile storage, as the client side have
no way to tell if the data is already written.
Submitted by: rmacklem
MFC after: 2 weeks
in r245004. Although the report was for noatime option which is
non-functional for the nullfs, other standard options like nosuid or
noexec are useful with it.
Reported by: Dewayne Geraghty <dewayne.geraghty@heuristicsystems.com.au>
MFC after: 3 days
by returning an error of EINTR rather than EACCES.
- While here, bring back some (but not all) of the NFS RPC statistics lost
when krpc was committed.
Reviewed by: rmacklem
MFC after: 1 week
existing nullfs vnode by the lower vnode is only 16 slots. Since the
default mode for the nullfs is to cache the vnodes, hash has extremely
huge chains.
Size the nullfs hashtbl based on the current value of
desiredvnodes. Use vfs_hash_index() to calculate the hash bucket for a
given vnode.
Pointy hat to: kib
Diagnosed and reviewed by: peter
Tested by: peter, pho (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 5 days
get back the leased write reference from the lower vnode. There is no
other path which can correct v_writecount on the lowervp.
Reported by: flo
Tested by: pho
Sponsored by: The FreeBSD Foundation
MFC after: 3 days
It was plagued with style errors and the offsets had been lost.
While here took the time to update the fields according to the
latest ext4 documentation.
Reviewed by: bde
MFC after: 3 days
Use file name hash as a tree key, handle duplicate keys. Both VOP_LOOKUP
and VOP_READDIR operations utilize same tree for search. Directory
entry offset (cookie) is either file name hash or incremental id in case
of hash collisions (duplicate-cookies). Keep sorted per directory list
of duplicate-cookie entries to facilitate cookie number allocation.
Don't fail if previous VOP_READDIR() offset is no longer valid, start
with next dirent instead. Other file system handle it similarly.
Workaround race prone tn_readdir_last[pn] fields update.
Add tmpfs_dir_destroy() to free all dirents.
Set NFS cookies in tmpfs_dir_getdents(). Return EJUSTRETURN from
tmpfs_dir_getdents() instead of hard coded -1.
Mark directory traversal routines static as they are no longer
used outside of tmpfs_subr.c
the free nullfs vnodes, switching nullfs behaviour to pre-r240285.
The option is mostly intended as the last-resort when higher pressure
on the vnode cache due to doubling of the vnode counts is not
desirable.
Note that disabling the cache costs more than 2x wall time in the
metadata-hungry scenarious. The default is "cache".
Tested and benchmarked by: pho (previous version)
MFC after: 2 weeks
We also try to make better use of the fs flags instead of
trying adapt the code according to the fs structures. In
the case of subsecond timestamps and birthtime we now
check that the feature is explicitly enabled: previously
we only checked that the reserved space was available and
silently wrote them.
This approach is much safer, especially if the filesystem
happens to use embedded inodes or support EAs.
Discussed with: Zheng Liu
MFC after: 5 days
to head. I don't think the NFS client behaviour will change unless
the new "minorversion=1" mount option is used. It includes basic
NFSv4.1 support plus support for pNFS using the Files Layout only.
All problems detecting during an NFSv4.1 Bakeathon testing event
in June 2012 have been resolved in this code and it has been tested
against the NFSv4.1 server available to me.
Although not reviewed, I believe that kib@ has looked at it.
which dumps out the actual options being used by an NFS mount.
This will be used to implement a "-m" option for nfsstat(1).
Reviewed by: alfred
MFC after: 2 weeks
Bring several definitions required for newer ext4 features.
Rename EXT2F_COMPAT_HTREE to EXT2F_COMPAT_DIRHASHINDEX since it
is not being used yet and the new name is more compatible with
NetBSD and Linux.
This change is purely cosmetic and has no effect on the real
code.
Obtained from: NetBSD
MFC after: 3 days
When a file is first being written, the dynamic block reallocation
(implemented by ext2_reallocblks) relocates the file's blocks
so as to cluster them together into a contiguous set of blocks on
the disk.
When the cluster crosses the boundary into the first indirect block,
the first indirect block is initially allocated in a position
immediately following the last direct block. Block reallocation
would usually destroy locality by moving the indirect block out of
the way to keep the data blocks contiguous.
The issue was diagnosed long ago by Bruce Evans on ffs and surfaced
on ext2fs when block reallocaton was ported. This is only a partial
solution based on the similarities with FFS. We still require more
review of the allocation details that vary in ext2fs.
Reported by: bde
MFC after: 1 week
the vnode use count, and this might cause the kernel to panic if compiled
with WITNESS enable.
- Be sure to put the '\0' terminator to the rpath string.
Sponsored by: iXsystems inc.
useful and has the side effect of obfuscating the code a bit.
- Remove spurious references to simple_lock.
Reported by: attilio [1]
Sponsored by: iXsystems inc.
vnode and following back the chain of n_parent pointers up to the root,
without acquiring the locks of the n_parent vnodes analyzed during the
computation. This is immediately wrong because if the vnode lock is not
held there's no guarantee on the validity of the vnode pointer or the data.
In order to fix, store the whole path in the smbnode structure so that
smbfs_fullpath() can use this information.
Discussed with: kib
Reported and tested by: pho
Sponsored by: iXsystems inc.
received granular locking) but the comment present in UFS has been
copied all over other filesystems code incorrectly for several times.
Removes comments that makes no sense now.
Reviewed by: kib
MFC after: 3 days
zombie list for the pid. This allows several kern.proc sysctls to
report useful information for zombies.
Hold the allproc_lock around all searches instead of relocking it.
Remove private pfind_locked() from the new nfs client code.
Requested and reviewed by: pjd
Tested by: pho
MFC after: 3 weeks
on timing of the operations and not real lookup, bringing too many
false positives. Remove the whole mechanism. If it needs to be
implemented, next time it should really be done in the proper way.
- Fix VOP_GETATTR() in order to cope with userland bugs that would
change the type of file and not panic. Instead it gets the entry as
if it is not existing.
Reported and tested by: flo
MFC after: 2 months
X-MFC: 241519, 242536,242616
was still possible to open for write from the lower filesystem. There
is a symmetric situation where the binary could already has file
descriptors opened for write, but it can be executed from the nullfs
overlay.
Handle the issue by passing one v_writecount reference to the lower
vnode if nullfs vnode has non-zero v_writecount. Note that only one
write reference can be donated, since nullfs only keeps one use
reference on the lower vnode. Always use the lower vnode v_writecount
for the checks.
Introduce the VOP_GET_WRITECOUNT to read v_writecount, which is
currently always bypassed to the lower vnode, and VOP_ADD_WRITECOUNT
to manipulate the v_writecount value, which manages a single bypass
reference to the lower vnode. Caling the VOPs instead of directly
accessing v_writecount provide the fix described in the previous
paragraph.
Tested by: pho
MFC after: 3 weeks
- Change the code so that it relies on vfs_hash rather than on a
home-made hashtable.
- There's no need to inline fnv_32_buf().
Reviewed by: delphij
Tested by: pho
Sponsored by: iXsystems inc.
VM pressure. The reason is that in some codepaths pointers to stack
variables were passed from one thread to another.
In collaboration with: pho
Reported by: pho's stress2 suite
Sponsored by: iXsystems inc.
- Use M_ZERO flag in malloc() rather than bzero()
- malloc() with M_NOWAIT can't return NULL so there's no need to check
Reviewed by: alc
Approved by: alc
In particular, do not lock Giant conditionally when calling into the
filesystem module, remove the VFS_LOCK_GIANT() and related
macros. Stop handling buffers belonging to non-mpsafe filesystems.
The VFS_VERSION is bumped to indicate the interface change which does
not result in the interface signatures changes.
Conducted and reviewed by: attilio
Tested by: pho
processes running as root to suspend/resume execution
of the kernel nfsd threads. An earlier version of this
patch was tested by Vincent Hoffman (vince at unsane.co.uk)
and John Hickey (jh at deterlab.net).
Reviewed by: kib
MFC after: 2 weeks
This has been developed during 2 summer of code mandates and being revived
by gnn recently.
The functionality in this commit mirrors entirely content of fusefs-kmod
port, which doesn't need to be installed anymore for -CURRENT setups.
In order to get some sparse technical notes, please refer to:
http://lists.freebsd.org/pipermail/freebsd-fs/2012-March/013876.html
or to the project branch:
svn://svn.freebsd.org/base/projects/fuse/
which also contains granular history of changes happened during port
refinements. This commit does not came from the branch reintegration
itself because it seems svn is not behaving properly for this functionaly
at the moment.
Partly Sponsored by: Google, Summer of Code program 2005, 2011
Originally submitted by: ilya, Csaba Henk <csaba-ml AT creo DOT hu >
In collabouration with: pho
Tested by: flo, gnn, Gustau Perez,
Kevin Oberman <rkoberman AT gmail DOT com>
MFC after: 2 months
If you have a binary on a filesystem which is also mounted over by
nullfs, you could execute the binary from the lower filesystem, or
from the nullfs mount. When executed from lower filesystem, the lower
vnode gets VV_TEXT flag set, and the file cannot be modified while the
binary is active. But, if executed as the nullfs alias, only the
nullfs vnode gets VV_TEXT set, and you still can open the lower vnode
for write.
Add a set of VOPs for the VV_TEXT query, set and clear operations,
which are correctly bypassed to lower vnode.
Tested by: pho (previous version)
MFC after: 2 weeks
and owner_group strings that consist entirely of
digits, interpreting them as the uid/gid number.
This change was needed since new (>= 3.3) Linux
servers reply with these strings by default.
This change is mandated by the rfc3530bis draft.
Reported on freebsd-stable@ under the Subject
heading "Problem with Linux >= 3.3 as NFSv4 server"
by Norbert Aschendorff on Aug. 20, 2012.
Tested by: norbert.aschendorff at yahoo.de
Reviewed by: jhb
MFC after: 2 weeks
which is an elaborate way to cause kernel panic. Change the VOPs
implementation to return EBADF for a reclaimed vnode.
While the calls to vop_bmap should not reach deadfs, it is indeed
possible for vop_ioctl, because the VOP locking protocol is to pass
the vnode to VOP unlocked. The actual panic was observed when ioctl
was called on procfs filedescriptor which pointed to an exited
process.
Reported by: zont
Tested by: pho
MFC after: 1 week
Use it for a printf() that can be harmlessly generated for mmap()'d
files. It will be used extensively for the NFSv4.1 client.
Debugging printf()s are enabled by setting vfs.nfs.debuglevel to
a non-zero value. The higher the value, the more debugging printf()s.
Reviewed by: jhb
MFC after: 2 weeks
it. There are two problems which shall be addressed for shared
lookups use to have measurable effect on nullfs scalability:
1. When vfs_lookup() calls VOP_LOOKUP() for nullfs, which passes lookup
operation to lower fs, resulting vnode is often only shared-locked. Then
null_nodeget() cannot instantiate covering vnode for lower vnode, since
insmntque1() and null_hashins() require exclusive lock on the lower.
Change the assert that lower vnode is exclusively locked to only
require any lock. If null hash failed to find pre-existing nullfs
vnode for lower vnode and the vnode is shared-locked, the lower vnode
lock is upgraded.
2. Nullfs reclaims its vnodes on deactivation. This is due to nullfs
inability to detect reclamation of the lower vnode. Reclamation of a
nullfs vnode at deactivation time prevents a reference to the lower
vnode to become stale.
Change nullfs VOP_INACTIVE to not reclaim the vnode, instead use the
VFS_RECLAIM_LOWERVP to get notification and reclaim upper vnode
together with the reclamation of the lower vnode.
Note that nullfs reclamation procedure calls vput() on the lowervp
vnode, temporary unlocking the vnode being reclaimed. This seems to be
fine for MPSAFE filesystems, but not-MPSAFE code often put partially
initialized vnode on some globally visible list, and later can decide
that half-constructed vnode is not needed. If nullfs mount is created
above such filesystem, then other threads might catch such not
properly initialized vnode. Instead of trying to overcome this case,
e.g. by recursing the lower vnode lock in null_reclaim_lowervp(), I
decided to rely on nearby removal of the support for non-MPSAFE
filesystems.
In collaboration with: pho
MFC after: 3 weeks
1) It is not useful to call "devfs_clear_cdevpriv()" from
"d_close" callbacks, hence for example read, write, ioctl and
so on might be sleeping at the time of "d_close" being called
and then then freed private data can still be accessed.
Examples: dtrace, linux_compat, ksyms (all fixed by this patch)
2) In sys/dev/drm* there are some cases in which memory will
be freed twice, if open fails, first by code in the open
routine, secondly by the cdevpriv destructor. Move registration
of the cdevpriv to the end of the drm open routines.
3) devfs_clear_cdevpriv() is not called if the "d_open" callback
registered cdevpriv data and the "d_open" callback function
returned an error. Fix this.
Discussed with: phk
MFC after: 2 weeks
network file systems (not only NFS proper). Short reads cause pages
other then the requested one, which were not filled by read response,
to stay invalid.
Change the vm_page_readahead_finish() interface to not take the error
code, but instead to make a decision to free or to (de)activate the
page only by its validity. As result, not requested invalid pages are
freed even if the read RPC indicated success.
Noted and reviewed by: alc
MFC after: 1 week
to pull vm_param.h was removed. Other big dependency of vm_page.h on
vm_param.h are PA_LOCK* definitions, which are only needed for
in-kernel code, because modules use KBI-safe functions to lock the
pages.
Stop including vm_param.h into vm_page.h. Include vm_param.h
explicitely for the kernel code which needs it.
Suggested and reviewed by: alc
MFC after: 2 weeks
vm_page oflags by providing helper function
vm_page_readahead_finish(), which handles completed reads for pages
with indexes other then the requested one, for VOP_GETPAGES().
Reviewed by: alc
MFC after: 1 week
I found 8.3 is a history BSD version using socket to implement FIFO
pipe, it uses per-file seqcount to compare with writer generation
stored in per-pipe object. The concept is after all writers are gone,
the pipe enters next generation, all old readers have not closed the
pipe should get the indication that the pipe is disconnected, result
is they should get EPIPE, SIGPIPE or get POLLHUP in poll().
But newcomer should not know that previous writters were gone, it
should treat it as a fresh session.
I am trying to bring back FIFO pipe to history behavior. It is still
unclear that if single EOF flag can represent SBS_CANTSENDMORE and
SBS_CANTRCVMORE which socket-based version is using, but I have run
the poll regression test in tool directory, output is same as the one
on 8.3-STABLE now.
I think the output "not ok 18 FIFO state 6b: poll result 0 expected 1.
expected POLLHUP; got 0" might be bogus, because newcomer should not
know that old writers were gone. I got the same behavior on Linux.
Our implementation always return POLLIN for disconnected pipe even it
should return POLLHUP, but I think it is not wise to remove POLLIN for
compatible reason, this is our history behavior.
Regression test: /usr/src/tools/regression/poll
flag but not PIPE_WANTW, but FIFO pipe code does not understand this internal
state, when a FIFO peer reader closes the pipe, it wants to notify the writer,
it checks PIPE_WANTW, if not set, it skips calling wakeup(), so blocked writer
never noticed the case, but in general, the writer should return from the
syscall with EPIPE error code and may get SIGPIPE signal. Setting the
PIPE_WANTW fixed problem, or you can turn off direct write, it should fix the
problem too. This bug is found by PR/170203.
Another bug in FIFO pipe code is when peer closes the pipe, another end which
is being blocked in select() or poll() is not notified, it missed to call
pipeselwakeup().
Third problem is found in poll regression test, the existing code can not
pass 6b,6c,6d tests, but FreeBSD-4 works. This commit does not fix the
problem, I still need to study more to find the cause.
PR: 170203
Tested by: Garrett Copper < yanegomi at gmail dot com >
now fully encapsulates all accesses to f_offset, and extends f_offset
locking to other consumers that need it, in particular, to lseek() and
variants of getdirentries().
Ensure that on 32bit architectures f_offset, which is 64bit quantity,
always read and written under the mtxpool protection. This fixes
apparently easy to trigger race when parallel lseek()s or lseek() and
read/write could destroy file offset.
The already broken ABI emulations, including iBCS and SysV, are not
converted (yet).
Tested by: pho
No objections from: jhb
MFC after: 3 weeks
bwrite(). VFS needs to know about EFAULT from uiomove() and does not
care much that partially filled block writeback after EFAULT was
successfull. Early return without error causes short write to be
reported to usermode.
Reported and tested by: andreast
MFC after: 3 weeks
written, but not msync'd by a process. A VOP_PUTPAGES()
called when VOP_RECLAIM() happens will usually fail, since
the NFSv4 Open has already been closed by VOP_INACTIVE().
Add a vm_object_page_clean() call to the NFSv4 client's
VOP_INACTIVE(), so that the write happens before the NFSv4
Open is closed. kib@ suggested using vgone() instead and
I will explore this, but this patch fixes things in the
meantime. For some reason, the VOP_PUTPAGES() is still
attaempted in VOP_RECLAIM(), but having this fail doesn't
cause any problems except a "stateid0 in write" being logged.
Reviewed by: kib
MFC after: 1 week
Do not brelse() the buffer unconditionally with BIO_ERROR set if
uiomove() failed. The brelse() treats most buffers with BIO_ERROR as
B_INVAL, dropping their content. Instead, if the write request
covered the whole buffer, remember the cached state and brelse() with
BIO_ERROR set only if the buffer was not cached previously.
Update the buffer dirtyoff/dirtyend based on the progress recorded by
uiomove() in passed struct uio, even in the presence of
error. Otherwise, usermode could see changed data in the backed pages,
but later the buffer is destroyed without write-back.
If uiomove() failed for IO_UNIT request, try to truncate the vnode
back to the pre-write state, and rewind the progress in passed uio
accordingly, following the FFS behaviour.
Reviewed by: rmacklem (some time ago)
Tested by: pho
MFC after: 1 month
Revamp the CAM enclosure services driver.
This updated driver uses an in-kernel daemon to track state changes and
publishes physical path location information\for disk elements into the
CAM device database.
Sponsored by: Spectra Logic Corporation
Sponsored by: iXsystems, Inc.
Submitted by: gibbs, will, mav
to freebsd-fs@, where the setfacl of an NFSv4 acl would fail.
This was caused by the VOP_ACLCHECK() call for ZFS replying
EOPNOTSUPP. After discussion with rwatson@, it was determined
that a call to VOP_ACLCHECK() before doing VOP_SETACL() is not
required. This patch fixes the problem by deleting the
VOP_ACLCHECK() call.
Tested by: Andrew Leonard (previous version)
MFC after: 1 week
The NAND Flash environment consists of several distinct components:
- NAND framework (drivers harness for NAND controllers and NAND chips)
- NAND simulator (NANDsim)
- NAND file system (NAND FS)
- Companion tools and utilities
- Documentation (manual pages)
This work is still experimental. Please use with caution.
Obtained from: Semihalf
Supported by: FreeBSD Foundation, Juniper Networks
8.x code:
- If the lock cannot be acquired immediately unlocks 'bar' vnode
and then locks both vnodes in order.
- wrong vnode type panics from cache_enter_time after calls by
ext2_lookup.
The fix merges the fixes from ufs/ufs_lookup.c.
Submitted by: Mateusz Guzik
Approved by: jhb@ (mentor)
Reviewed by: kib@
MFC after: 1 week
