| Age | Commit message (Collapse) | Author |
|
Forgot what I was debugging when I inserted a relookup in my local
tree months or years ago, but whatever it was, if that solved a
problem, this KDASSERT will make the problem more obvious.
|
|
size dependant on memory size. If less than 128MB of memory, default
to no cache. With 128MB of memory or more, use a maximum cache size of
1/64th of memory; cap maximum default cache size to 32MB (for systems
with 2GB of memory or more).
The dirhash cache sizes are still explicityly setable by sysctl(8) or
by adding relevant entry(s) to sysctl.conf(5).
|
|
|
|
|
|
|
|
|
|
|
|
Remove unused predicates vn_locked() and vn_anylocked().
Welcome to 9.99.95
|
|
- Centralize vnode kevent handling in the VOP_*() wrappers, rather than
forcing each individual file system to deal with it (except VOP_RENAME(),
because VOP_RENAME() is a mess and we currently have 2 different ways
of handling it; at least it's reasonably well-centralized in the "new"
way).
- Add support for NOTE_OPEN, NOTE_CLOSE, NOTE_CLOSE_WRITE, and NOTE_READ,
compatible with the same events in FreeBSD.
- Track which kevent notifications clients are interested in receiving
to avoid doing work for events no one cares about (avoiding, e.g.
taking locks and traversing the klist to send a NOTE_WRITE when
someone is merely watching for a file to be deleted, for example).
In support of the above:
- Add support in vnode_if.sh for specifying PRE- and POST-op handlers,
to be invoked before and after vop_pre() and vop_post(), respectively.
Basic idea from FreeBSD, but implemented differently.
- Add support in vnode_if.sh for specifying CONTEXT fields in the
vop_*_args structures. These context fields are used to convey information
between the file system VOP function and the VOP wrapper, but do not
occupy an argument slot in the VOP_*() call itself. These context fields
are initialized and subsequently interpreted by PRE- and POST-op handlers.
- Version VOP_REMOVE(), uses the a context field for the file system to report
back the resulting link count of the target vnode. Return this in tmpfs,
udf, nfs, chfs, ext2fs, lfs, and ufs.
NetBSD 9.99.92.
|
|
|
|
|
|
|
|
|
|
These are things of the form #define foofs_op genfs_op, or #define
foofs_op genfs_eopnotsupp, or similar. They serve no purpose besides
obfuscation, and have gotten cutpasted all over everywhere.
|
|
Add GENFS_SPECOP_ENTRIES and GENFS_FIFOOP_ENTRIES macros that contain
the portion of the vnode ops table declaration that is
(conservatively) the same in every fs. Use these in every fs that
supports devices and/or fifos with separate ops tables.
Note that ptyfs works differently (it has one type of vnode with
open-coded dispatch to the specfs code, which I haven't changed in
this commit) and rump/librump/rumpvfs/rumpfs.c has an indirect dynamic
dispatch that already does more or less the same thing, which I also
haven't changed.
Also note that this anticipates a few bits in the next changeset here
and there, and adds missing but unreachable calls in some cases (e.g.
most fses weren't defining whiteout on devices and fifos, but it isn't
reachable there), and it changes parsepath on devices and fifos to
genfs_badop from genfs_parsepath (but it's not reachable there
either).
It appears that devices in kernfs were missing kqfilter, so it's
possible that if you try to use kqueue on /kern/rootdev that it'll
explode.
And finally note that the ops declaration tables aren't
order-dependent. (Other than vop_default_desc has to come first.)
Otherwise this wouldn't work.
|
|
Cloning devices (and also things like /dev/stderr) work by allocating
a struct file, stuffing it in the file table (which is a layer
violation), stuffing the file descriptor number for it in a magic
field of struct lwp (which is gross), and then "failing" with one of
two magic errnos, EDUPFD or EMOVEFD.
Before this commit, all callers of vn_open in the kernel (there are
quite a few) were expected to check for these errors and handle the
situation. Needless to say, none of them except for open() itself did,
resulting in internal negative errnos being returned to userspace.
This hack is fairly deeply rooted and cannot be eliminated all at
once. This commit adds logic to handle the magic errnos inside
vn_open; now on success vn_open returns either a vnode or an integer
file descriptor, along with a flag that says whether the underlying
code requested EDUPFD or EMOVEFD. Callers not prepared to cope with
file descriptors can pass NULL for the extra return values, in which
case if a file descriptor would be produced vn_open fails with
EOPNOTSUPP.
Since I'm rearranging vn_open's signature anyway, stop exposing struct
nameidata. Instead, take three arguments: an optional vnode to use as
the starting point (like openat()), the path, and additional namei
flags to use, restricted to NOCHROOT and TRYEMULROOT. (Other namei
behavior, e.g. NOFOLLOW, can be requested via the open flags.)
This change requires a kernel bump. Ride the one an hour ago.
(That was supposed to be coordinated; did not intend to let an hour
slip by. My fault.)
|
|
- Move namei_getcomponent to genfs_vnops.c and call it genfs_parsepath.
- Add a parsepath entry to every vnode ops table.
VOP_PARSEPATH takes a directory vnode to be searched and a complete
following path and chooses how much of that path to consume. To begin
with, all parsepath calls are genfs_parsepath, which locates the first
'/' as always.
Note that the call doesn't take the whole struct componentname, only
the string. The other bits of struct componentname should not be
needed and there's no reason to cause potential complications by
exposing them.
|
|
|
|
We can remove -Wno-error=address-of-packed-member from various
makefiles now.
|
|
The poorly named uvm.h is generally supposed to be for uvm-internal
users only.
- Narrow it to files that actually need it -- mostly files that need
to query whether curlwp is the pagedaemon, which should maybe be
exposed by an external header.
- Use uvm_extern.h where feasible and uvm_*.h for things not exposed
by it. We should split up uvm_extern.h but this will serve for now
to reduce the uvm.h dependencies.
- Use uvm_stat.h and #ifdef UVMHIST uvm.h for files that use
UVMHIST(ubchist), since ubchist is declared in uvm.h but the
reference evaporates if UVMHIST is not defined, so we reduce header
file dependencies.
- Make uvm_device.h and uvm_swap.h independently includable while
here.
ok chs@
|
|
This change made no sense and should not have been committed.
|
|
Fix some missing uvm_vnp_setsize in screw cases while here.
|
|
They no longer matter on disk so we don't need to write anything out
for them.
|
|
|
|
cached value will do, or if the very latest total must be fetched. It can
be called thousands of times a second and fetching the totals impacts not
only the calling LWP but other CPUs doing unrelated activity in the VM
system.
|
|
|
|
itself)
- Add new flag UBC_ISMAPPED which tells ubc_uiomove() the object is mmap()ed
somewhere. Use it to decide whether to do direct-mapped copy, rather than
poking around directly in the vnode in ubc_uiomove(), which is ugly and
doesn't work for tmpfs. It would be nicer to contain all this in UVM but
the filesystem provides the needed locking here (VV_MAPPED) and to
reinvent that would suck more.
- Rename UBC_UNMAP_FLAG() to UBC_VNODE_FLAGS(). Pass in UBC_ISMAPPED where
appropriate.
|
|
that hides the details and does atomic_load_relaxed(). Signature matches
FreeBSD.
|
|
and getcwd():
- push vnode locking back as far as possible.
- do most lookups directly in the namecache, avoiding vnode locks & refs.
- don't block new refs to vnodes across VOP_INACTIVE().
- get shared locks for VOP_LOOKUP() if the file system supports it.
- correct lock types for VOP_ACCESS() / VOP_GETATTR() in a few places.
Possible future enhancements:
- make the lookups lockless.
- support dotdot lookups by being lockless and inferring absence of chroot.
- maybe make it work for layered file systems.
- avoid vnode references at the root & cwd.
|
|
|
|
lfs64 directory entries are only 32-bit aligned in order to conserve
space in directory blocks, and we had a hack to stuff a 64-bit inode
in them. This replaces the hack by __aligned(4) __packed, and goes
further:
1. It's not clear that all the other lfs64 data structures are 64-bit
aligned on disk to begin with. We can go through these later and
upgrade them from
struct foo64 {
...
} __aligned(4) __packed;
union foo {
struct foo64 f64;
...
};
to
struct foo64 {
...
};
union foo {
struct foo64 f64 __aligned(8);
...
} __aligned(4) __packed;
if we really want to take advantage of 64-bit memory accesses.
However, the __aligned(4) __packed must remain on the union
because:
2. We access even the lfs32 data structures via a union that has
lfs64 members, and it turns out that compilers will assume access
through a union with 64-bit aligned members implies the whole
union has 64-bit alignment, even if we're only accessing a 32-bit
aligned member.
|
|
|
|
Remove unneeded changes and only deal with the PQ_WANTED flag, to exclude
possible bugs.
|
|
automate installation of sysctl nodes.
Note that there are still a number of device and pseudo-device modules
that create entries tied to individual device units, rather than to the
module itself. These are not changed.
|
|
work list. Proposed on tech-kern@.
|
|
locks out of the equation for sleep/wakeup, and allows observing+waiting
for busy pages when holding only a read lock. Proposed on tech-kern.
|
|
functions: preempt_point() and preempt_needed().
- preempt(): if the LWP has exceeded its timeslice in kernel, strip it of
any priority boost gained earlier from blocking.
|
|
|
|
- Change the lock on uvm_object, vm_amap and vm_anon to be a RW lock.
- Break v_interlock and vmobjlock apart. v_interlock remains a mutex.
- Do partial PV list locking in the x86 pmap. Others to follow later.
|
|
Just wanted to confirm that a race might happen, and indeed it did.
These serve little diagnostic value otherwise.
|
|
|
|
|
|
|
|
Assert that mtsleep can't fail either -- it doesn't catch signals and
there's no timeout.
|
|
|
|
Otherwise, we might do
lfs_vflush
-> lfs_seglock
-> lfs_segwait(SEGM_CKP)
-> lfs_writer_enter
which is the reverse of the lfs_writer -> lfs_seglock ordering.
|
|
Fixes reloading lfs.kmod.
|
|
The lock order is lfs_writer -> lfs_seglock. The problem in 52301 is
that lfs_segwrite violates this lock order by sometimes doing
lfs_seglock -> lfs_writer, either (a) when doing a checkpoint or (b),
opportunistically, when there are no dirops pending. Both cases can
deadlock, because dirops sometimes take the seglock (lfs_truncate,
lfs_valloc, lfs_vfree):
(a) There may be dirops pending, and they may be waiting for the
seglock, so we can't wait for them to complete while holding the
seglock.
(b) The test for fs->lfs_dirops == 0 happens unlocked, and the state
may change by the time lfs_writer_enter acquires lfs_lock.
To resolve this in each case:
(a) Do lfs_writer_enter before lfs_seglock, since we will need it
unconditionally anyway. The worst performance impact of this should
be that some dirops get delayed a little bit.
(b) Create a new lfs_writer_tryenter to use at this point so that the
test for fs->lfs_dirops == 0 and the acquisition of lfs_writer happen
atomically under lfs_lock.
|
|
There's no need to lfs_writer_enter at all here, as far as I can see.
lfs_flush_fs will do it for us.
|
|
Fixes panic:
KASSERT((ip->i_state & IN_ADIROP) == 0) at lfs_vnops.c:1670
lfs_flush_dirops
lfs_check
lfs_setattr
VOP_SETATTR
change_mode
sys_fchmod
syscall
This assertion -- and the assertion that vp->v_uflag has VU_DIROP set
-- is valid only until we release lfs_lock, because we may race with
lfs_unmark_dirop which will remove the nodes and change the flags.
Further, vp itself is valid only as long as it is referenced, which it
is as long as it's on the dchain, but lfs_unmark_dirop drops the
dchain's reference.
|