MAKEDEV: Add MAKEDEV glue for the ti(4) device nodes.
ti.4: Update the ti(4) man page to include information on the
TI_JUMBO_HDRSPLIT and TI_PRIVATE_JUMBOS kernel options,
and also include information about the new character
device interface and the associated ioctls.
man9/Makefile: Add jumbo.9 and zero_copy.9 man pages and associated
links.
jumbo.9: New man page describing the jumbo buffer allocator
interface and operation.
zero_copy.9: New man page describing the general characteristics of
the zero copy send and receive code, and what an
application author should do to take advantage of the
zero copy functionality.
NOTES: Add entries for ZERO_COPY_SOCKETS, TI_PRIVATE_JUMBOS,
TI_JUMBO_HDRSPLIT, MSIZE, and MCLSHIFT.
conf/files: Add uipc_jumbo.c and uipc_cow.c.
conf/options: Add the 5 options mentioned above.
kern_subr.c: Receive side zero copy implementation. This takes
"disposable" pages attached to an mbuf, gives them to
a user process, and then recycles the user's page.
This is only active when ZERO_COPY_SOCKETS is turned on
and the kern.ipc.zero_copy.receive sysctl variable is
set to 1.
uipc_cow.c: Send side zero copy functions. Takes a page written
by the user and maps it copy on write and assigns it
kernel virtual address space. Removes copy on write
mapping once the buffer has been freed by the network
stack.
uipc_jumbo.c: Jumbo disposable page allocator code. This allocates
(optionally) disposable pages for network drivers that
want to give the user the option of doing zero copy
receive.
uipc_socket.c: Add kern.ipc.zero_copy.{send,receive} sysctls that are
enabled if ZERO_COPY_SOCKETS is turned on.
Add zero copy send support to sosend() -- pages get
mapped into the kernel instead of getting copied if
they meet size and alignment restrictions.
uipc_syscalls.c:Un-staticize some of the sf* functions so that they
can be used elsewhere. (uipc_cow.c)
if_media.c: In the SIOCGIFMEDIA ioctl in ifmedia_ioctl(), avoid
calling malloc() with M_WAITOK. Return an error if
the M_NOWAIT malloc fails.
The ti(4) driver and the wi(4) driver, at least, call
this with a mutex held. This causes witness warnings
for 'ifconfig -a' with a wi(4) or ti(4) board in the
system. (I've only verified for ti(4)).
ip_output.c: Fragment large datagrams so that each segment contains
a multiple of PAGE_SIZE amount of data plus headers.
This allows the receiver to potentially do page
flipping on receives.
if_ti.c: Add zero copy receive support to the ti(4) driver. If
TI_PRIVATE_JUMBOS is not defined, it now uses the
jumbo(9) buffer allocator for jumbo receive buffers.
Add a new character device interface for the ti(4)
driver for the new debugging interface. This allows
(a patched version of) gdb to talk to the Tigon board
and debug the firmware. There are also a few additional
debugging ioctls available through this interface.
Add header splitting support to the ti(4) driver.
Tweak some of the default interrupt coalescing
parameters to more useful defaults.
Add hooks for supporting transmit flow control, but
leave it turned off with a comment describing why it
is turned off.
if_tireg.h: Change the firmware rev to 12.4.11, since we're really
at 12.4.11 plus fixes from 12.4.13.
Add defines needed for debugging.
Remove the ti_stats structure, it is now defined in
sys/tiio.h.
ti_fw.h: 12.4.11 firmware.
ti_fw2.h: 12.4.11 firmware, plus selected fixes from 12.4.13,
and my header splitting patches. Revision 12.4.13
doesn't handle 10/100 negotiation properly. (This
firmware is the same as what was in the tree previously,
with the addition of header splitting support.)
sys/jumbo.h: Jumbo buffer allocator interface.
sys/mbuf.h: Add a new external mbuf type, EXT_DISPOSABLE, to
indicate that the payload buffer can be thrown away /
flipped to a userland process.
socketvar.h: Add prototype for socow_setup.
tiio.h: ioctl interface to the character portion of the ti(4)
driver, plus associated structure/type definitions.
uio.h: Change prototype for uiomoveco() so that we'll know
whether the source page is disposable.
ufs_readwrite.c:Update for new prototype of uiomoveco().
vm_fault.c: In vm_fault(), check to see whether we need to do a page
based copy on write fault.
vm_object.c: Add a new function, vm_object_allocate_wait(). This
does the same thing that vm_object allocate does, except
that it gives the caller the opportunity to specify whether
it should wait on the uma_zalloc() of the object structre.
This allows vm objects to be allocated while holding a
mutex. (Without generating WITNESS warnings.)
vm_object_allocate() is implemented as a call to
vm_object_allocate_wait() with the malloc flag set to
M_WAITOK.
vm_object.h: Add prototype for vm_object_allocate_wait().
vm_page.c: Add page-based copy on write setup, clear and fault
routines.
vm_page.h: Add page based COW function prototypes and variable in
the vm_page structure.
Many thanks to Drew Gallatin, who wrote the zero copy send and receive
code, and to all the other folks who have tested and reviewed this code
over the years.
This facilitates the use in circumstances where you are using a serial
console as well. GDB doesn't support anything higher than 9600 baud (19k2
if you are lucky), but the console does.
options do. Comments should be in NOTES and having the comments in two
places usually means that one place will just bitrot. Thus, remove the
comment for KTRACE_REQUEST_POOL from the previous revision.
Requested by: bde
operations to dump a ktrace event out to an output file are now handled
asychronously by a ktrace worker thread. This enables most ktrace events
to not need Giant once p_tracep and p_traceflag are suitably protected by
the new ktrace_lock.
There is a single todo list of pending ktrace requests. The various
ktrace tracepoints allocate a ktrace request object and tack it onto the
end of the queue. The ktrace kernel thread grabs requests off the head of
the queue and processes them using the trace vnode and credentials of the
thread triggering the event.
Since we cannot assume that the user memory referenced when doing a
ktrgenio() will be valid and since we can't access it from the ktrace
worker thread without a bit of hassle anyways, ktrgenio() requests are
still handled synchronously. However, in order to ensure that the requests
from a given thread still maintain relative order to one another, when a
synchronous ktrace event (such as a genio event) is triggered, we still put
the request object on the todo list to synchronize with the worker thread.
The original thread blocks atomically with putting the item on the queue.
When the worker thread comes across an asynchronous request, it wakes up
the original thread and then blocks to ensure it doesn't manage to write a
later event before the original thread has a chance to write out the
synchronous event. When the original thread wakes up, it writes out the
synchronous using its own context and then finally wakes the worker thread
back up. Yuck. The sychronous events aren't pretty but they do work.
Since ktrace events can be triggered in fairly low-level areas (msleep()
and cv_wait() for example) the ktrace code is designed to use very few
locks when posting an event (currently just the ktrace_mtx lock and the
vnode interlock to bump the refcoun on the trace vnode). This also means
that we can't allocate a ktrace request object when an event is triggered.
Instead, ktrace request objects are allocated from a pre-allocated pool
and returned to the pool after a request is serviced.
The size of this pool defaults to 100 objects, which is about 13k on an
i386 kernel. The size of the pool can be adjusted at compile time via the
KTRACE_REQUEST_POOL kernel option, at boot time via the
kern.ktrace_request_pool loader tunable, or at runtime via the
kern.ktrace_request_pool sysctl.
If the pool of request objects is exhausted, then a warning message is
printed to the console. The message is rate-limited in that it is only
printed once until the size of the pool is adjusted via the sysctl.
I have tested all kernel traces but have not tested user traces submitted
by utrace(2), though they should work fine in theory.
Since a ktrace request has several properties (content of event, trace
vnode, details of originating process, credentials for I/O, etc.), I chose
to drop the first argument to the various ktrfoo() functions. Currently
the functions just assume the event is posted from curthread. If there is
a great desire to do so, I suppose I could instead put back the first
argument but this time make it a thread pointer instead of a vnode pointer.
Also, KTRPOINT() now takes a thread as its first argument instead of a
process. This is because the check for a recursive ktrace event is now
per-thread instead of process-wide.
Tested on: i386
Compiles on: sparc64, alpha
is currently conditional on both the GEOM and GEOM_GPT options to
avoid getting GPT by default and having the MBR and GPT classes
clash.
The correct behaviour of the MBR class would be to back-off (reject)
a MBR if it's a Protective MBR (a MBR with a single partition of type
0xEE that spans the whole disk (as far as the MBR is concerned).
The correct behaviour if the GPT class would be to back-off (reject)
a GPT if there's a MBR that's not a Protective MBR.
At this stage it's inconvenient to destroy a good MBR when working
with GPTs that it's more convenient to have the MBR class back-off
when it detects the GPT signature on disk and have the GPT class
ignore the MBR.
In sys/gpt.h UUIDs (GUIDs) for the following FreeBSD partitions
have been defined:
GPT_ENT_TYPE_FREEBSD
FreeBSD slice with disklabel. This is the equivalent of
the well-known FreeBSD MBR partition type.
GPT_ENT_TYPE_FREEBSD_{SWAP|UFS|UFS2|VINUM}
FreeBSD partitions in the context of disklabel. This is
speculating on the idea to use the GPT to hold partitions
instead if slices and removing the fixed (and low) limits
we have on the number of partitions.
This commit lacks a GPT image for the regression suite.
option is used (not on by default).
- In the case of trying to lock a mutex, if the MTX_CONTESTED flag is set,
then we can safely read the thread pointer from the mtx_lock member while
holding sched_lock. We then examine the thread to see if it is currently
executing on another CPU. If it is, then we keep looping instead of
blocking.
- In the case of trying to unlock a mutex, it is now possible for a mutex
to have MTX_CONTESTED set in mtx_lock but to not have any threads
actually blocked on it, so we need to handle that case. In that case,
we just release the lock as if MTX_CONTESTED was not set and return.
- We do not adaptively spin on Giant as Giant is held for long times and
it slows SMP systems down to a crawl (it was taking several minutes,
like 5-10 or so for my test alpha and sparc64 SMP boxes to boot up when
they adaptively spinned on Giant).
- We only compile in the code to do this for SMP kernels, it doesn't make
sense for UP kernels.
Tested on: i386, alpha, sparc64
IFS had its fingers deep in the belly of the UFS/FFS split. IFS
will be reimplemented by the maintainer at a later date.
Requested by: adrian (maintainer)
behavior by default. Also, change the options line to reflect this.
If there are no problems reported this will become the only behavior and the
knob will be removed in a month or so.
Demanded by: obrien
following sysctl variables:
debug.mutex.prof.enable enable / disable profiling
debug.mutex.prof.acquisitions number of mutex acquisitions recorded
debug.mutex.prof.records number of acquisition points recorded
debug.mutex.prof.maxrecords max number of acquisition points
debug.mutex.prof.rejected number of rejections (due to full table)
debug.mutex.prof.hashsize hash size
debug.mutex.prof.collisions number of hash collisions
debug.mutex.prof.stats profiling statistics
The code records four numbers for each acquisition point (identified by
source file name and line number): longest time held, total time held,
number of non-recursive acquisitions, average time held. The measurements
are in clock cycles (as returned by get_cyclecount(9)); this may cause
measurements on some SMP systems to be unreliable. This can probably be
worked around by replacing get_cyclecount(9) by some incarnation of
nanotime(9).
This work was derived from initial patches by eivind.
dump the trace buffer feasible.
- Remove KTR_EXTEND. This changes the format of the trace entries when
activated, making writing a userland tool which is not tied to a specific
kernel configuration difficult.
- Use get_cyclecount() for timestamps. nanotime() is much too heavy weight
and requires recursion protection due to ktr traces occuring as a result
of ktr traces. KTR_VERBOSE may still require recursion protection, which
is now conditional on it.
- Allow KTR_CPU to be overridden by MD code. This is so that it is possible
to trace early in startup before pcpu and/or curthread are setup.
- Add a version number for the ktr interface. A userland tool can check this
to detect mismatches.
- Use an array for the parameters to make decoding in userland easier.
- Add file and line recording to the non-extended traces now that the extended
version is no more.
These changes will break gdb macros to decode the extended version of the
trace buffer which are floating around. Users of these macros should either
use the show ktr command in ddb, or use the userland utility which can be run
on a core dump.
Approved by: jhb
Tested on: i386, sparc64
The stat() and open() calls have been changed to make use of this new functionality. Using shared locks in
these cases is sufficient and can significantly reduce their latency if IO is pending to these vnodes. Also,
this reduces the number of exclusive locks that are floating around in the system, which helps reduce the
number of deadlocks that occur.
A new kernel option "LOOKUP_SHARED" has been added. It defaults to off so this patch can be turned on for
testing, and should eventually go away once it is proven to be stable. I have personally been running this
patch for over a year now, so it is believed to be fully stable.
Reviewed by: jake, obrien
Approved by: jake
- Temporary fix a bug of Intel ACPI CA core code.
- Add OS layer ACPI mutex support. This can be disabled by
specifying option ACPI_NO_SEMAPHORES.
- Add ACPI threading support. Now that we have a dedicate taskqueue for
ACPI tasks and more ACPI task threads can be created by specifying option
ACPI_MAX_THREADS.
- Change acpi_EvaluateIntoBuffer() behavior slightly to reuse given
caller's buffer unless AE_BUFFER_OVERFLOW occurs. Also CM battery's
evaluations were changed to use acpi_EvaluateIntoBuffer().
- Add new utility function acpi_ConvertBufferToInteger().
- Add simple locking for CM battery and temperature updating.
- Fix a minor problem on EC locking.
- Make the thermal zone polling rate to be changeable.
- Change minor things on AcpiOsSignal(); in ACPI_SIGNAL_FATAL case,
entering Debugger is easier to investigate the problem rather than panic.
built without support for miibus PHYs. Most ed cards don't need
miibus support, so it's useful to be able to avoid the bloat of
all the mii devices for small fixed-purpose kernels.
then one can restart from a panic by resetting the panicstr variable to
NULL. This commit conditionalizes the previously committed functionality
on this variable. It also removes the __dead2 attribute from the panic()
function so that when one continues from a panic() the behavior will
be predictable.
information. The default limits only effect machines with > 1GB of ram
and can be overriden with two new kernel conf variables VM_SWZONE_SIZE_MAX
and VM_BCACHE_SIZE_MAX, or with loader variables kern.maxswzone and
kern.maxbcache. This has the effect of leaving more KVM available for
sizing NMBCLUSTERS and 'maxusers' and should avoid tripups where a sysad
adds memory to a machine and then sees the kernel panic on boot due to
running out of KVM.
Also change the default swap-meta auto-sizing calculation to allocate half
of what it was previously allocating. The prior defaults were way too high.
Note that we cannot afford to run out of swap-meta structures so we still
stay somewhat conservative here.
directories. When enabled via "options UFS_DIRHASH", in-core hash
arrays are maintained for large directories. These allow all
directory operations to take place quickly instead of requiring
long linear searches. For now anyway, dirhash is not enabled by
default.
The in-core hash arrays have a memory requirement that is approximately
half the size of the size of the on-disk directory file. A number
of new sysctl variables allow control over which directories get
hashed and over the maximum amount of memory that dirhash will use:
vfs.ufs.dirhash_minsize
The minimum on-disk directory size for which hashing should be
used. The default is 2560 (2.5k).
vfs.ufs.dirhash_maxmem
The system-wide maximum total memory to be used by dirhash data
structures. The default is 2097152 (2MB).
The current amount of memory being used by dirhash is visible
through the read-only sysctl variable vfs.ufs.dirhash_maxmem.
Finally, some extra sanity checks that are enabled by default, but
which may have an impact on performance, can be disabled by setting
vfs.ufs.dirhash_docheck to 0.
Discussed on: -fs, -hackers
1. Add SA_IO_TIMEOUT as an option (4 minutes default) to cover reads,
writes, wfm, test unit ready.
2. Add internal SCSIOP_TIMEOUT (e.g., for mode sense) at 1 minute. This
should not require an option, but is cleaner to parameterize.
MFC after: 1 week
This closes a minor information leak which allows a remote observer to
determine the rate at which the machine is generating packets, since the
default behaviour is to increment a counter for each packet sent.
Reviewed by: -net
Obtained from: OpenBSD
