kld_unload event handler which gets invoked after a linker file has been
successfully unloaded. The kld_unload and kld_load event handlers are now
invoked with the shared linker lock held, while kld_unload_try is invoked
with the lock exclusively held.
Convert hwpmc(4) to use these event handlers instead of having
kern_kldload() and kern_kldunload() invoke hwpmc(4) hooks whenever files are
loaded or unloaded. This has no functional effect, but simplifes the linker
code somewhat.
Reviewed by: jhb
with rmlocks. This works only with non-sleepable rm because handlers run
in SWI context. While here, document the new KPI in the timeout(9)
manpage.
Requested by: adrian, scottl
Reviewed by: mav, remko(manpage)
existing examples to not pass an mbuf as a probe argument. There's no
obvious reason to have it there, and it doesn't really jibe with the example
added in this revision.
MFC after: 1 week
called after the module has been loaded, and the unload handlers are called
before the module is unloaded. Moreover, the module unload handlers may
return an error to prevent the unload from proceeding.
Reviewed by: avg
MFC after: 2 weeks
Now the MTX_RECURSE flag can be passed to the mtx_*_flag() calls.
This helps in cases we want to narrow down to specific calls the
possibility to recurse for some locks.
Sponsored by: EMC / Isilon storage division
Reviewed by: jeff, alc
Tested by: pho
Unify the 2 concept into a real, minimal, sxlock where the shared
acquisition represent the soft busy and the exclusive acquisition
represent the hard busy.
The old VPO_WANTED mechanism becames the hard-path for this new lock
and it becomes per-page rather than per-object.
The vm_object lock becames an interlock for this functionality:
it can be held in both read or write mode.
However, if the vm_object lock is held in read mode while acquiring
or releasing the busy state, the thread owner cannot make any
assumption on the busy state unless it is also busying it.
Also:
- Add a new flag to directly shared busy pages while vm_page_alloc
and vm_page_grab are being executed. This will be very helpful
once these functions happen under a read object lock.
- Move the swapping sleep into its own per-object flag
The KPI is heavilly changed this is why the version is bumped.
It is very likely that some VM ports users will need to change
their own code.
Sponsored by: EMC / Isilon storage division
Discussed with: alc
Reviewed by: jeff, kib
Tested by: gavin, bapt (older version)
Tested by: pho, scottl
some general word-smithing.
- Don't claim that adaptive mutexes have a timeout (they don't).
- Don't treat pool mutexes as a separate primitive in a few places.
- Describe sleepable read-mostly locks as a separate lock type and add
them to the various tables.
- Don't claim that sx locks are less efficient. That hasn't been true in
a few years now.
- Describe lockmanager locks next to sx locks since they are very similar
in terms of rules, etc., and so that all the lock primitives are
grouped together before the non-lock primitives.
- Similarly, move the section on Giant after the description of all the
non-lock primitives to preserve grouping.
- Condition variables work on several types of locks, not just mutexes.
- Add a bit of language to compare/contrast condition variables with
sleep/wakeup.
- Add a note about why pause(9) is unique.
- Add some language to define bounded vs unbounded sleeps and explain
why they are treated separately (bounded sleeps only need CPU time
to make forward progress).
- Don't state that using mtx_sleep() is a bad idea. It is in fact rather
necessary.
- Rework the interaction table a bit. First, it did not include really
include sleepable rmlocks and it left out lockmgr entirely. To get
things to fit, combine similar lock types into the same column / row,
and explicitly state what "sleep" means. The notes about recursion
and lock order were also a bit banal (lock order is always important,
not just in the few places annotated here), so remove them. In
particular, the lock order note would need to be on just about every
cell. If we want to document recursion I think a better approach
would be a separate table summarizing the recursion rules for each
lock as having too many notes clutters the table.
- Tweak the tables to use less indentation so everything still fits with
the added columns.
- Correct a few cells in the context mode table.
- Use mdoc markup instead of explicit markup in a few places.
Requested by: julian
MFC after: 2 weeks
provided by Isilon.
- Add an rm_assert() supporting various lock assertions similar to other
locking primitives. Because rmlocks track readers the assertions are
always fully accurate unlike rw_assert() and sx_assert().
- Flesh out the lock class methods for rmlocks to support sleeping via
condvars and rm_sleep() (but only while holding write locks), rmlock
details in 'show lock' in DDB, and the lc_owner method used by
dtrace.
- Add an internal destroyed cookie so that API functions can assert
that an rmlock is not destroyed.
- Make use of rm_assert() to add various assertions to the API (e.g.
to assert locks are held when an unlock routine is called).
- Give RM_SLEEPABLE locks their own lock class and always use the
rmlock's own lock_object with WITNESS.
- Use THREAD_NO_SLEEPING() / THREAD_SLEEPING_OK() to disallow sleeping
while holding a read lock on an rmlock.
Submitted by: andre
Obtained from: EMC/Isilon
Introduce counter(9) API, that implements fast and raceless counters,
provided (but not limited to) for gathering of statistical data.
See http://lists.freebsd.org/pipermail/freebsd-arch/2013-April/014204.html
for more details.
In collaboration with: kib
Reviewed by: luigi
Tested by: ae, ray
Sponsored by: Nginx, Inc.
These zones have slab size == sizeof(struct pcpu), but request from VM
enough pages to fit (uk_slabsize * mp_ncpus). An item allocated from such
zone would have a separate twin for each CPU in the system, and these twins
are at a distance of sizeof(struct pcpu) from each other. This magic value
of distance would allow us to make some optimizations later.
To address private item from a CPU simple arithmetics should be used:
item = (type *)((char *)base + sizeof(struct pcpu) * curcpu)
These arithmetics are available as zpcpu_get() macro in pcpu.h.
To introduce non-page size slabs a new field had been added to uma_keg
uk_slabsize. This shifted some frequently used fields of uma_keg to the
fourth cache line on amd64. To mitigate this pessimization, uma_keg fields
were a bit rearranged and least frequently used uk_name and uk_link moved
down to the fourth cache line. All other fields, that are dereferenced
frequently fit into first three cache lines.
Sponsored by: Nginx, Inc.
The scope of these callbacks is primarily to support actions that affect the
taskqueue's thread environments. They are entirely optional, and
consequently are introduced as a new API: taskqueue_set_callback().
This interface allows the caller to specify that a taskqueue requires a
callback and optional context pointer for a given callback type.
The callback types included in this commit can be used to register a
constructor and destructor for thread-local storage using osd(9). This
allows a particular taskqueue to define that its threads require a specific
type of TLS, without the need for a specially-orchestrated task-based
mechanism for startup and shutdown in order to accomplish it.
Two callback types are supported at this point:
- TASKQUEUE_CALLBACK_TYPE_INIT, called by every thread when it starts, prior
to processing any tasks.
- TASKQUEUE_CALLBACK_TYPE_SHUTDOWN, called by every thread when it exits,
after it has processed its last task but before the taskqueue is
reclaimed.
While I'm here:
- Add two new macros, TQ_ASSERT_LOCKED and TQ_ASSERT_UNLOCKED, and use them
in appropriate locations.
- Fix taskqueue.9 to mention taskqueue_start_threads(), which is a required
interface for all consumers of taskqueue(9).
Reviewed by: kib (all), eadler (taskqueue.9), brd (taskqueue.9)
Approved by: ken (mentor)
Sponsored by: Spectra Logic
MFC after: 1 month
Switch eventtimers(9) from using struct bintime to sbintime_t.
Even before this not a single driver really supported full dynamic range of
struct bintime even in theory, not speaking about practical inexpediency.
This change legitimates the status quo and cleans up the code.
in the man page and its header counterpart.
Submitted by: Christoph Mallon <christoph.mallon@gmx.de> (initial version)
Reviewed and further improved by: bde (previous version)
All bugs are: mine
will be printed once the given zone becomes full and cannot allocate an
item. The warning will not be printed more often than every five minutes.
All UMA warnings can be globally turned off by setting sysctl/tunable
vm.zone_warnings to 0.
Discussed on: arch
Obtained from: WHEEL Systems
MFC after: 2 weeks
Since SMPng it wasn't really supported anymore and if it worked
then only by chance. Only very few drivers ever supported it.
Discussed with: yongari
MFC after: 2 weeks
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
filters (ipfw and PF) use the same ruleset with the same lock for both
AF_INET and AF_INET6 there is no need in more fine-grade locking.
However, it is possible to request personal lock by specifying
PFIL_FLAG_PRIVATE_LOCK flag in pfil_head structure (see pfil.9 for
more details).
Export PFIL lock via rw_lock(9)/rm_lock(9)-like API permitting pfil consumers
to use this lock instead of own lock. This help reducing locks on main
traffic path.
pfil_assert() is currently not implemented due to absense of rm_assert().
Waiting for some kind of r234648 to be merged in HEAD.
This change is part of bigger patch reducing routing locking.
Sponsored by: Yandex LLC
Reviewed by: glebius, ae
OK'd by: silence on net@
MFC after: 3 weeks
- All packets in NETISR_IP queue are in net byte order.
- ip_input() is entered in net byte order and converts packet
to host byte order right _after_ processing pfil(9) hooks.
- ip_output() is entered in host byte order and converts packet
to net byte order right _before_ processing pfil(9) hooks.
- ip_fragment() accepts and emits packet in net byte order.
- ip_forward(), ip_mloopback() use host byte order (untouched actually).
- ip_fastforward() no longer modifies packet at all (except ip_ttl).
- Swapping of byte order there and back removed from the following modules:
pf(4), ipfw(4), enc(4), if_bridge(4).
- Swapping of byte order added to ipfilter(4), based on __FreeBSD_version
- __FreeBSD_version bumped.
- pfil(9) manual page updated.
Reviewed by: ray, luigi, eri, melifaro
Tested by: glebius (LE), ray (BE)
tree used it incorrectly, which lead to inaccurate overrated
if_obytes accounting. The drbr(9) used to update ifnet stats on
drbr_enqueue(), which is not accurate since enqueuing doesn't
imply successful processing by driver. Dequeuing neither mean
that. Most drivers also called drbr_stats_update() which did
accounting again, leading to doubled if_obytes statistics. And
in case of severe transmitting, when a packet could be several
times enqueued and dequeued it could have been accounted several
times.
o Thus, make drbr(9) API thinner. Now drbr(9) merely chooses between
ALTQ queueing or buf_ring(9) queueing.
- It doesn't touch the buf_ring stats any more.
- It doesn't touch ifnet stats anymore.
- drbr_stats_update() no longer exists.
o buf_ring(9) handles its stats itself:
- It handles br_drops itself.
- br_prod_bytes stats are dropped. Rationale: no one ever
reads them but update of a common counter on every packet
negatively affects performance due to excessive cache
invalidation.
- buf_ring_enqueue_bytes() reduced to buf_ring_enqueue(), since
we no longer account bytes.
o Drivers handle their stats theirselves: if_obytes, if_omcasts.
o mlx4(4), igb(4), em(4), vxge(4), oce(4) and ixv(4) no longer
use drbr_stats_update(), and update ifnet stats theirselves.
o bxe(4) was the most correct driver, it didn't call
drbr_stats_update(), thus it was the only driver accurate under
moderate load. Now it also maintains stats itself.
o ixgbe(4) had already taken stats from hardware, so just
- drop software stats updating.
- take multicast packet count from hardware as well.
o mxge(4) just no longer needs NO_SLOW_STATS define.
o cxgb(4), cxgbe(4) need no change, since they obtain stats
from hardware.
Reviewed by: jfv, gnn
- Provide missing function that can do hashing of arbitrary sized buffer.
- Refetch lookup3.c and do only minimal edits to it, so that diff between
our jenkins_hash.c and lookup3.c is minimal.
- Add declarations for jenkins_hash(), jenkins_hash32() to sys/hash.h.
- Document these functions in hash(9)
Obtained from: http://burtleburtle.net/bob/c/lookup3.c
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
a da(4) instance going away while GEOM is still probing it.
In this case, the GEOM disk class instance has been created by
disk_create(), and the taste of the disk is queued in the GEOM
event queue.
While that event is queued, the da(4) instance goes away. When the
open call comes into the da(4) driver, it dereferences the freed
(but non-NULL) peripheral pointer provided by GEOM, which results
in a panic.
The solution is to add a callback to the GEOM disk code that is
called when all of its resources are cleaned up. This is
implemented inside GEOM by adding an optional callback that is
called when all consumers have detached from a provider, and the
provider is about to be deleted.
scsi_cd.c,
scsi_da.c: In the register routine for the cd(4) and da(4)
routines, acquire a reference to the CAM peripheral
instance just before we call disk_create().
Use the new GEOM disk d_gone() callback to register
a callback (dadiskgonecb()/cddiskgonecb()) that
decrements the peripheral reference count once GEOM
has finished cleaning up its resources.
In the cd(4) driver, clean up open and close
behavior slightly. GEOM makes sure we only get one
open() and one close call, so there is no need to
set an open flag and decrement the reference count
if we are not the first open.
In the cd(4) driver, use cam_periph_release_locked()
in a couple of error scenarios to avoid extra mutex
calls.
geom.h: Add a new, optional, providergone callback that
is called when a provider is about to be deleted.
geom_disk.h: Add a new d_gone() callback to the GEOM disk
interface.
Bump the DISK_VERSION to version 2. This probably
should have been done after a couple of previous
changes, especially the addition of the d_getattr()
callback.
geom_disk.c: Add a providergone callback for the disk class,
g_disk_providergone(), that calls the user's
d_gone() callback if it exists.
Bump the DISK_VERSION to 2.
geom_subr.c: In g_destroy_provider(), call the providergone
callback if it has been provided.
In g_new_geomf(), propagate the class's
providergone callback to the new geom instance.
blkfront.c: Callers of disk_create() are supposed to pass in
DISK_VERSION, not an explicit disk API version
number. Update the blkfront driver to do that.
disk.9: Update the disk(9) man page to include information
on the new d_gone() callback, as well as the
previously added d_getattr() callback, d_descr
field, and HBA PCI ID fields.
MFC after: 5 days
- Consistently refer to rmlocks as "read-mostly locks".
- Relate rmlocks to rwlocks rather than sx locks since they are closer to
rwlocks.
- Add a separate paragraph on sleepable read-mostly locks contrasting them
with "normal" read-mostly locks.
- The flag passed to rm_init_flags() to enable recursion for readers is
RM_RECURSE, not LO_RECURSABLE.
- Fix the description for RM_RECURSE (it allows readers to recurse, not
writers).
- Explicitly note that rm_try_rlock() honors RM_RECURSE.
- Fix some minor grammar nits.
Fix panic on tcpdump being attached to interface being removed (introduced by r233937, pointed by hrs@ and adrian@)
Protect most of bpf_setf() by BPF global lock
Add several forgotten assertions (thanks to adrian@)
Document current locking model inside bpf.c
Document EVENTHANDLER(9) usage inside BPF.
Approved by: kib(mentor)
Tested by: gnn
MFC in: 4 weeks
Function acquired reader lock if needed.
Assert check for reader or writer lock (RA_LOCKED / RA_UNLOCKED)
- While here, add knlist_init_mtx.9 to MLINKS and fix some style(9) issues
Reviewed by: glebius
Approved by: ae(mentor)
MFC after: 2 weeks
3rd argument of ifa->ifa_rtrequest is now ``rt_addrinfo *'' instead of
``sockaddr *''. While here, un-document RTM_RESOLVE cmd argument for
ifa_rtrequest() that became a stub after separating L2 tables in r186119.
MFC after: 1 week
- Document the following routines: pci_alloc_msi(), pci_alloc_msix(),
pci_find_cap(), pci_get_max_read_req(), pci_get_vpd_ident(),
pci_get_vpd_readonly(), pci_msi_count(), pci_msix_count(),
pci_pending_msix(), pci_release_msi(), pci_remap_msix(), and
pci_set_max_read_req().
- Group the functions into five sub-sections: raw configuration access,
locating devices, device information, device configuration, and
message signaled interrupts.
- Discourage use of pci_disable_io() and pci_enable_io() in device drivers.
The PCI bus driver handles this automatically as resources are activated.
MFC after: 2 weeks
not ACPI-specific at all, but deal with PCI power states. Also,
pci_set_powerstate() fails with EOPNOTSUPP if a request is made that the
underlying device does not support rather than falling back to somehow
setting D0.
pci_cfg_save() and pci_cfg_restore() for device drivers to use when
saving and restoring state (e.g. to handle device-specific resets).
Reviewed by: imp
MFC after: 2 weeks
The reasoning behind this, is that if we are consistent in our
documentation about the uint*_t stuff, people will be less tempted to
write new code that uses the non-standard types.
I am not going to bump the man page dates, as these changes can be
considered style nits. The meaning of the man pages is unaffected.
MFC after: 1 month
These functions take a `struct cdev *' -- not a dev_t. Inside the
kernel, dev_t has the same use as in userspace, namely to store a device
identifier.
MFC after: 2 weeks
I was considering adding it to libc as well, but last minute I thought
it would be good enough to add it to libkern exclusively. I forgot to
rename the man page and hook it up.
It seems two of the file system drivers we have in the tree, namely ufs
and ext3, use a function called `skpc()'. The meaning of this function
does not seem to be documented in FreeBSD, but it turns out one needs to
be a VAX programmer to understand what it does.
SPKC is an instruction on the VAX that does the opposite of memchr(). It
searches for the non-equal character. Add a new function called
memcchr() to the tree that has the following advantages over skpc():
- It has a name that makes more sense than skpc(). Just like strcspn()
matches the complement of strspn(), memcchr() is the complement of
memchr().
- It is faster than skpc(). Similar to our strlen() in libc, it compares
entire words, instead of single bytes. It seems that for this routine
this yields a sixfold performance increase on amd64.
- It has a man page.
objects created by shm_open(2) into the kernel's address space. This
provides a convenient way for creating shared memory buffers between
userland and the kernel without requiring custom character devices.
and DEVMETHOD() we can fully hide the explicit mention of kobj(9) from
device drivers.
- Update the example in driver.9 to use DEVMETHOD_END.
Submitted by: jhb
MFC after: 3 days
This enables locking consumers to pass their own structures around as const and
be able to assert locks embedded into those structures.
Reviewed by: ed, kib, jhb
curthread-accessing part of mtx_{,un}lock(9) when using a r210623-style
curthread implementation on sparc64, crashing the kernel in its early
cycles as PCPU isn't set up, yet (and can't be set up as OFW is one of the
things we need for that, which leads to a chicken-and-egg problem). What
happens is that due to the fact that the idea of r210623 actually is to
allow the compiler to cache invocations of curthread, it factors out
obtaining curthread needed for both mtx_lock(9) and mtx_unlock(9) to
before the branch based on kobj_mutex_inited when compiling the kernel
without the debugging options. So change kobj_class_compile_static(9)
to just never acquire kobj_mtx, effectively restricting it to its
documented use, and add a kobj_init_static(9) for initializing objects
using a class compiled with the former and that also avoids using mutex(9)
(and malloc(9)). Also assert in both of these functions that they are
used in their intended way only.
While at it, inline kobj_register_method() and kobj_unregister_method()
as there wasn't much point for factoring them out in the first place
and so that a reader of the code has to figure out the locking for
fewer functions missing a KOBJ_ASSERT.
Tested on powerpc{,64} by andreast.
Reviewed by: nwhitehorn (earlier version), jhb
MFC after: 3 days
document knlist_delete, and better document what knlist_clear does... Note
that both of these functions may sleep, and also unlock/relock the list
lock...
document knlist_init_mtx (forgotten by kib)...
other minor improvements
Reviewed by: ru (previous rev)
MFC after: 1 week
