There are several negative side-effects of not calling into VOP layer
at all for page cache reads. The biggest is the missed activation of
EVFILT_READ knotes.
Also, it allows filesystem to make more fine grained decision to
refuse read from page cache.
Keep VIRF_PGREAD flag around, it is still useful for nullfs, and for
asserts.
Reviewed by: markj
Tested by: pho
Discussed with: mjg
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D26346
The current scheme of calling VOP_GETATTR adds avoidable overhead.
An example with tmpfs doing fstat (ops/s):
before: 7488958
after: 7913833
Reviewed by: kib (previous version)
Differential Revision: https://reviews.freebsd.org/D25910
The only reason to vlazy there is to (overzealously) ensure all vnodes
which need to be visited by msync scan can be found there.
In particluar this is of no use zfs and tmpfs.
While here depessimize the check.
This obviates the need to scan the entire active list looking for vnodes
of interest.
msync is handled by adding all vnodes with write count to the lazy list.
deferred inactive directly adds vnodes as it sets the VI_DEFINACT flag.
Vnodes get dequeued from the list when their hold count reaches 0.
Newly added MNT_VNODE_FOREACH_LAZY* macros support filtering so that
spurious locking is avoided in the common case.
Reviewed by: jeff
Tested by: pho (in a larger patch, previous version)
Differential Revision: https://reviews.freebsd.org/D22995
Filesystems which want to use it in limited capacity can employ the
VOP_UNLOCK_FLAGS macro.
Reviewed by: kib (previous version)
Differential Revision: https://reviews.freebsd.org/D21427
r355677 modified the NFS client so that it does lseek(SEEK_DATA/SEEK_HOLE)
for NFSv4.2, but calls vop_stdioctl() otherwise. As such, vop_stdioctl()
needs to be a global function.
Missed during the code merge for r355677.
Both of these features are not needed by many consumers and result in avoidable
reads which in turn puts them on profiles due to cache-line ping ponging.
On top of that the current lockgmr entry point is slower than necessary
single-threaded. As an attempted clean up preparing for other changes,
provide new routines which don't support any of the aforementioned features.
With these patches in place vop_stdlock and vop_stdunlock disappear from
flamegraphs during -j 104 buildkernel.
Reviewed by: jeff (previous version)
Tested by: pho
Differential Revision: https://reviews.freebsd.org/D22665
The current vnode layout is not smp-friendly by having frequently read data
avoidably sharing cachelines with very frequently modified fields. In
particular v_iflag inspected for VI_DOOMED can be found in the same line with
v_usecount. Instead make it available in the same cacheline as the v_op, v_data
and v_type which all get read all the time.
v_type is avoidably 4 bytes while the necessary data will easily fit in 1.
Shrinking it frees up 3 bytes, 2 of which get used here to introduce a new
flag field with a new value: VIRF_DOOMED.
Reviewed by: kib, jeff
Differential Revision: https://reviews.freebsd.org/D22715
This one is very hard to run into. If the filesystem is being unmounted or
the mount point is freed the vfs_op_thread_enter will fail. For it to
succeed the mount point itself would have to be reallocated in the time
window between the initial read and the attempt to enter.
Sponsored by: The FreeBSD Foundation
The breakage was added after all the testing and the testing which followed
was not sufficient to find it.
Reported by: pho
Sponsored by: The FreeBSD Foundation
There are 3 counters modified all the time in this structure - one for
keeping the structure alive, one for preventing unmount and one for
tracking active writers. Exact values of these counters are very rarely
needed, which makes them a prime candidate for conversion to a per-cpu
scheme, resulting in much better performance.
Sample benchmark performing fstatfs (modifying 2 out of 3 counters) on
a 104-way 2 socket Skylake system:
before: 852393 ops/s
after: 76682077 ops/s
Reviewed by: kib, jeff
Tested by: pho
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21637
New primitive is introduced to denote sections can operate locklessly
on aspects of struct mount, but which can also be disabled if necessary.
This provides an opportunity to start scaling common case modifications
while providing stable state of the struct when facing unmount, write
suspendion or other events.
mnt_ref is the first counter to start being managed in this manner with
the intent to make it per-cpu.
Reviewed by: kib, jeff
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21425
The initially read mount point can already be NULL.
Reported by: markj
Fixes: r351656 ("vfs: stop refing freed mount points in vop_stdgetwritemount")
Sponsored by: The FreeBSD Foundation
The code used blindly ref based on an unsafely red address and then would
backpedal if necessary. This was safe in terms of memory access since
mounts are type-stable, but made for a potential a bug where the mount
was reused and had the count reset to 0 before this code decreased it.
Reviewed by: kib
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21411
vnode usecount drops to 0 all the time (e.g. for directories during path lookup).
When that happens the kernel would always lock the exclusive lock for the vnode
in order to call vinactive(). This blocks other threads who want to use the vnode
for looukp.
vinactive is very rarely needed and can be tested for without the vnode lock held.
This patch gives filesytems an opportunity to do it, sample total wait time for
tmpfs over 500 minutes of poudriere -j 104:
before: 557563641706 (lockmgr:tmpfs)
after: 46309603301 (lockmgr:tmpfs)
Sponsored by: The FreeBSD Foundation
Differential Revision: https://reviews.freebsd.org/D21371
Without this patch, when an application performed lseek(SEEK_DATA/SEEK_HOLE)
on a file in a file system that does not have its own VOP_IOCTL(), the
lseek(2) fails with errno ENOTTY. This didn't seem appropriate, since
ENOTTY is not listed as an error return by either the lseek(2) man page
nor the POSIX draft for lseek(2).
A discussion on freebsd-current@ seemed to indicate that implementing
a trivial algorithm that returns the offset argument for FIOSEEKDATA and
returns the file's size for FIOSEEKHOLE was the preferred fix.
http://docs.FreeBSD.org/cgi/mid.cgi?CAOtMX2iiQdv1+15e1N_r7V6aCx_VqAJCTP1AW+qs3Yg7sPg9wA
The Linux kernel appears to implement this trivial algorithm as well.
This patch adds a vop_stdioctl() that implements this trivial algorithm.
It returns errors consistent with vn_bmap_seekhole() and, as such, will
still return ENOTTY for non-regular files.
I have proposed a separate patch that maps errors not described by the
lseek(2) man page nor POSIX draft to EINVAL. This patch is under separate
review.
Reviewed by: kib
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D21299
Suppose that a binary was executed from tmpfs mount, and the text
vnode was reclaimed while the binary was still running. It is
possible during even the normal operations since tmpfs vnode'
vm_object has swap type, and no references on the vnode is held. Also
assume that the text vnode was revived for some reason. Then, on the
process exit or exec, unmapping of the text mapping tries to remove
the text reference from the vnode, but since it went from
recycle/instantiation cycle, there is no reference kept, and assertion
in VOP_UNSET_TEXT_CHECKED() triggers.
Fix this by keeping a use reference on the tmpfs vnode for each exec
reference. This prevents the vnode reclamation while executable map
entry is active.
Do it by adding per-mount flag MNTK_TEXT_REFS that directs
vop_stdset_text() to add use ref on first vnode text use, and
per-vnode VI_TEXT_REF flag, to record the need on unref in
vop_stdunset_text() on last vnode text use going away. Set
MNTK_TEXT_REFS for tmpfs mounts.
Reported by: bdrewery
Tested by: sbruno, pho (previous version)
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
This patch adds support to the kernel for a Linux compatible
copy_file_range(2) syscall and the related VOP_COPY_FILE_RANGE(9).
This syscall/VOP can be used by the NFSv4.2 client to implement the
Copy operation against an NFSv4.2 server to do file copies locally on
the server.
The vn_generic_copy_file_range() function in this patch can be used
by the NFSv4.2 server to implement the Copy operation.
Fuse may also me able to use the VOP_COPY_FILE_RANGE() method.
vn_generic_copy_file_range() attempts to maintain holes in the output
file in the range to be copied, but may fail to do so if the input and
output files are on different file systems with different _PC_MIN_HOLE_SIZE
values.
Separate commits will be done for the generated syscall files and userland
changes. A commit for a compat32 syscall will be done later.
Reviewed by: kib, asomers (plus comments by brooks, jilles)
Relnotes: yes
Differential Revision: https://reviews.freebsd.org/D20584
The order is correct, it is nullfs vnode interlock -> lower vnode
interlock. vop_stdadd_writecount() is called from nullfs
VOP_ADD_WRITECOUNT() and both take interlocks.
Requested by: markj
Sponsored by: The FreeBSD Foundation
MFC after: 2 weeks
kern_execve() locks text vnode exclusive to be able to set and clear
VV_TEXT flag. VV_TEXT is mutually exclusive with the v_writecount > 0
condition.
The change removes VV_TEXT, replacing it with the condition
v_writecount <= -1, and puts v_writecount under the vnode interlock.
Each text reference decrements v_writecount. To clear the text
reference when the segment is unmapped, it is recorded in the
vm_map_entry backed by the text file as MAP_ENTRY_VN_TEXT flag, and
v_writecount is incremented on the map entry removal
The operations like VOP_ADD_WRITECOUNT() and VOP_SET_TEXT() check that
v_writecount does not contradict the desired change. vn_writecheck()
is now racy and its use was eliminated everywhere except access.
Atomic check for writeability and increment of v_writecount is
performed by the VOP. vn_truncate() now increments v_writecount
around VOP_SETATTR() call, lack of which is arguably a bug on its own.
nullfs bypasses v_writecount to the lower vnode always, so nullfs
vnode has its own v_writecount correct, and lower vnode gets all
references, since object->handle is always lower vnode.
On the text vnode' vm object dealloc, the v_writecount value is reset
to zero, and deadfs vop_unset_text short-circuit the operation.
Reclamation of lowervp always reclaims all nullfs vnodes referencing
lowervp first, so no stray references are left.
Reviewed by: markj, trasz
Tested by: mjg, pho
Sponsored by: The FreeBSD Foundation
MFC after: 1 month
Differential revision: https://reviews.freebsd.org/D19923
Provide a convenience function to avoid the hack with filling fake
struct vop_fsync_args and then calling vop_stdfsync().
Sponsored by: The FreeBSD Foundation
MFC after: 1 week
POSIX_FADV_DONTNEED). The most broken case was for applications that
advise for the whole file and then do block-aligned i/o's 1 block at
a time. Then advice is sent to VOP_ADVISE() 1 block at a time, but
in vop_stdadvise() the 1-block advice was turned into 0-block advice
for the buffer cache part.
The bugs were caused partly by callers representing the region as
(a_start, a_end), where a_end is actually the maximum, and everything
else representing the region as (start, end) where 'end' is actually
the end (1 after the maximum). The maximum a_end must be rounded up,
but was rounded down. Also, rounding to page boundaries was inconsistent.
The bugs and fixes have no effect for zfs and other file systems that
don't use the buffer cache or the page cache. Most or all file systems
currently use the default VOP_FADVISE(), but it finds a null buffer cache
and a null page cache for file systems that don't use normal methods.
Reviewed by: kib
Use bypass to catch any NFS VOP dispatch and route it through the
wrapper which does sigdeferstop() and then dispatches original
VOP. NFS does not need a bypass below it, which is not supported.
The vop offset in the vop_vector is added since otherwise it is
impossible to get vop_op_t from the internal table, and I did not
wanted to create the layered fs only to wrap NFS VOPs.
VFS_OP()s wrap is straightforward.
Requested and reviewed by: mjg (previous version)
Tested by: pho
Sponsored by: The FreeBSD Foundation
Differential revision: https://reviews.freebsd.org/D17658
The interlock pointer is non-NULL by definition and the compiler see through
that and eliminates the NULL checks. Just remove them from the code as they
play no role.
No difference in generated assembly.
appeared on UFS/FFS filesystems. In some cases it was promptly followed
by a panic of "softdep_deallocate_dependencies: dangling deps". This fix
should eliminate both of these occurences.
Submitted by: Andreas Longwitz <longwitz at incore.de>
Reviewed by: kib
Tested by: Peter Holm (pho)
PR: 225423
MFC after: 1 week
On the one hand, FIFOs should respect other variables not supported by
the fifofs vnode operation (such as _PC_NAME_MAX, _PC_LINK_MAX, etc.).
These values are fs-specific and must come from a fs-specific method.
On the other hand, filesystems that support FIFOs are required to
support _PC_PIPE_BUF on directory vnodes that can contain FIFOs.
Given this latter requirement, once the fs-specific VOP_PATHCONF
method supports _PC_PIPE_BUF for directories, it is also suitable for
FIFOs permitting a single VOP_PATHCONF method to be used for both
FIFOs and non-FIFOs.
To that end, retire all of the FIFO-specific pathconf methods from
filesystems and change FIFO-specific vnode operation switches to use
the existing fs-specific VOP_PATHCONF method. For fifofs, set it's
VOP_PATHCONF to VOP_PANIC since it should no longer be used.
While here, move _PC_PIPE_BUF handling out of vop_stdpathconf() so that
only filesystems supporting FIFOs will report a value. In addition,
only report a valid _PC_PIPE_BUF for directories and FIFOs.
Discussed with: bde
Reviewed by: kib (part of a larger patch)
MFC after: 1 month
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D12572
Having all filesystems fall through to default values isn't always correct
and these values can vary for different filesystem implementations. Most
of these changes just use the existing default values with a few exceptions:
- Don't report CHOWN_RESTRICTED for ZFS since it doesn't do the exact
permissions check this claims for chown().
- Use NANDFS_NAME_LEN for NAME_MAX for nandfs.
- Don't report a LINK_MAX of 0 on smbfs. Now fail with EINVAL to
indicate hard links aren't supported.
Requested by: bde (though perhaps not this exact implementation)
Reviewed by: kib (earlier version)
MFC after: 1 month
Sponsored by: Chelsio Communications
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
Move handling of these three pathconf() variables out of vop_stdpathconf()
and into devfs_pathconf() as TTY devices can only be devfs files. In
addition, only return settings for these three variables for devfs devices
whose device switch has the D_TTY flag set.
Discussed with: bde, kib
Sponsored by: Chelsio Communications
reviewing all uses of OFF_TO_IDX(), I observed that
vm_object_page_noreuse() is requiring an exclusive lock on the object
when, in fact, a shared lock suffices.
Reviewed by: kib, markj
MFC after: 3 days
Differential Revision: https://reviews.freebsd.org/D10011
then return EAGAIN. The current code just returns that if the LAST buf
failed.
Reviewed by: kib@, trasz@
Differential Revision: https://reviews.freebsd.org/D9677
The main lockmgr routine takes 8 arguments which makes it impossible to
tail-call it by the intermediate vop_stdlock/unlock routines.
The routine itself starts with an if-forest and reads from the lock itself
several times.
This slows things down both single- and multi-threaded. With the patch
single-threaded fstats go 4% up and multithreaded up to ~27%.
Note that there is still a lot of room for improvement.
Reviewed by: kib
Tested by: pho
