The ability to schedule multiple threads per process
(one one cpu) by making ALL system calls optionally asynchronous.
to come: ia64 and power-pc patches, patches for gdb, test program (in tools)
Reviewed by: Almost everyone who counts
(at various times, peter, jhb, matt, alfred, mini, bernd,
and a cast of thousands)
NOTE: this is still Beta code, and contains lots of debugging stuff.
expect slight instability in signals..
out-of-range, drop the file reference before returning. (This error
also exists in the RELENG_4 branch.)
o Eliminate the acquisition and release of Giant in readv()
now that malloc() and free() are callable without Giant.
most cases NULL is passed, but in some cases such as network driver locks
(which use the MTX_NETWORK_LOCK macro) and UMA zone locks, a name is used.
Tested on: i386, alpha, sparc64
This shrinks the size 4 bytes on alpha, down to the same 276 bytes
as all other platforms.
Construct a hack to make old ioctls work on new kernels.
Once world is recompiled only the new and correct sysctls will be
used.
This hack will become annoying around 1st of may to make people
rebuild their worlds and it will be gone before 5.0.
kern/kern_descrip.c:
Aquire Giant in fdrop_locked when file refcount hits zero, this removes
the requirement for the caller to own Giant for the most part.
kern/kern_ktrace.c:
Aquire Giant in ktrgenio, simplifies locking in upper read/write syscalls.
kern/vfs_bio.c:
Aquire Giant in bwillwrite if needed.
kern/sys_generic.c
Giant pushdown, remove Giant for:
read, pread, write and pwrite.
readv and writev aren't done yet because of the possible malloc calls
for iov to uio processing.
kern/sys_socket.c
Grab giant in the socket fo_read/write functions.
kern/vfs_vnops.c
Grab giant in the vnode fo_read/write functions.
Problem:
selwakeup required calling pfind which would cause lock order
reversals with the allproc_lock and the per-process filedesc lock.
Solution:
Instead of recording the pid of the select()'ing process into the
selinfo structure, actually record a pointer to the thread. To
avoid dereferencing a bad address all the selinfo structures that
are in use by a thread are kept in a list hung off the thread
(protected by sellock). When a selwakeup occurs the selinfo is
removed from that threads list, it is also removed on the way out
of select or poll where the thread will traverse its list removing
all the selinfos from its own list.
Problem:
Previously the PROC_LOCK was used to provide the mutual exclusion
needed to ensure proper locking, this couldn't work because there
was a single condvar used for select and poll and condvars can
only be used with a single mutex.
Solution:
Introduce a global mutex 'sellock' which is used to provide mutual
exclusion when recording events to wait on as well as performing
notification when an event occurs.
Interesting note:
schedlock is required to manipulate the per-thread TDF_SELECT
flag, however if given its own field it would not need schedlock,
also because TDF_SELECT is only manipulated under sellock one
doesn't actually use schedlock for syncronization, only to protect
against corruption.
Proc locks are no longer used in select/poll.
Portions contributed by: davidc
other threads as well as speed up the interfaces.
To fix the race and accomplish the speedup, remove selholddrop and
pollholddrop. The entire concept is somewhat bogus because holding
the individual struct file pointers offers us no guarantees that
another thread context won't close it on us thereby removing our
access to our own reference.
Selholddrop and pollholddrop also would do multiple locks and unlocks
of mutexes _per-file_ in the fd arrays to be scanned, this needed to
be sped up.
Instead of using selholddrop and pollholddrop, simply hold the
filedesc lock over the selscan and pollscan functions. This should
protect us against close(2)'s on the files as reduce the multiple
lock/unlock pairs per fd into a single lock over the filedesc.
Seigo Tanimura (tanimura) posted the initial delta.
I've polished it quite a bit reducing the need for locking and
adapting it for KSE.
Locks:
1 mutex in each filedesc
protects all the fields.
protects "struct file" initialization, while a struct file
is being changed from &badfileops -> &pipeops or something
the filedesc should be locked.
1 mutex in each struct file
protects the refcount fields.
doesn't protect anything else.
the flags used for garbage collection have been moved to
f_gcflag which was the FILLER short, this doesn't need
locking because the garbage collection is a single threaded
container.
could likely be made to use a pool mutex.
1 sx lock for the global filelist.
struct file * fhold(struct file *fp);
/* increments reference count on a file */
struct file * fhold_locked(struct file *fp);
/* like fhold but expects file to locked */
struct file * ffind_hold(struct thread *, int fd);
/* finds the struct file in thread, adds one reference and
returns it unlocked */
struct file * ffind_lock(struct thread *, int fd);
/* ffind_hold, but returns file locked */
I still have to smp-safe the fget cruft, I'll get to that asap.
Replace uses of holdfp() with fget*() or fgetvp*() calls as appropriate
introduce fget(), fget_read(), fget_write() - these functions will take
a thread and file descriptor and return a file pointer with its ref
count bumped.
introduce fgetvp(), fgetvp_read(), fgetvp_write() - these functions will
take a thread and file descriptor and return a vref()'d vnode.
*_read() requires that the file pointer be FREAD, *_write that it be
FWRITE.
This continues the cleanup of struct filedesc and struct file access
routines which, when are all through with it, will allow us to then
make the API calls MP safe and be able to move Giant down into the fo_*
functions.
was locked by the proc lock and td_flags is locked by the sched_lock.
The places that read, set, and cleared TDF_SELECT weren't updated, so they
read and modified td_flags w/o holding the sched_lock, meaning that they
could corrupt the per-thread flags field. As an immediate band-aid,
grab sched_lock while reading and manipulating td_flags in relation to
TDF_SELECT. This will probably be cleaned up some later on.
Note ALL MODULES MUST BE RECOMPILED
make the kernel aware that there are smaller units of scheduling than the
process. (but only allow one thread per process at this time).
This is functionally equivalent to teh previousl -current except
that there is a thread associated with each process.
Sorry john! (your next MFC will be a doosie!)
Reviewed by: peter@freebsd.org, dillon@freebsd.org
X-MFC after: ha ha ha ha
- Since polling should not involve sleeping, keep holding a
process lock upon scanning file descriptors.
- Hold a reference to every file descriptor prior to entering
polling loop in order to avoid lock order reversal between
lockmgr and p_mtx upon calling fdrop() in fo_poll().
(NOTE: this work has not been done for netncp and netsmb
yet because a socket itself has no reference counts.)
Reviewed by: jhb
mtx_enter(lock, type) becomes:
mtx_lock(lock) for sleep locks (MTX_DEF-initialized locks)
mtx_lock_spin(lock) for spin locks (MTX_SPIN-initialized)
similarily, for releasing a lock, we now have:
mtx_unlock(lock) for MTX_DEF and mtx_unlock_spin(lock) for MTX_SPIN.
We change the caller interface for the two different types of locks
because the semantics are entirely different for each case, and this
makes it explicitly clear and, at the same time, it rids us of the
extra `type' argument.
The enter->lock and exit->unlock change has been made with the idea
that we're "locking data" and not "entering locked code" in mind.
Further, remove all additional "flags" previously passed to the
lock acquire/release routines with the exception of two:
MTX_QUIET and MTX_NOSWITCH
The functionality of these flags is preserved and they can be passed
to the lock/unlock routines by calling the corresponding wrappers:
mtx_{lock, unlock}_flags(lock, flag(s)) and
mtx_{lock, unlock}_spin_flags(lock, flag(s)) for MTX_DEF and MTX_SPIN
locks, respectively.
Re-inline some lock acq/rel code; in the sleep lock case, we only
inline the _obtain_lock()s in order to ensure that the inlined code
fits into a cache line. In the spin lock case, we inline recursion and
actually only perform a function call if we need to spin. This change
has been made with the idea that we generally tend to avoid spin locks
and that also the spin locks that we do have and are heavily used
(i.e. sched_lock) do recurse, and therefore in an effort to reduce
function call overhead for some architectures (such as alpha), we
inline recursion for this case.
Create a new malloc type for the witness code and retire from using
the M_DEV type. The new type is called M_WITNESS and is only declared
if WITNESS is enabled.
Begin cleaning up some machdep/mutex.h code - specifically updated the
"optimized" inlined code in alpha/mutex.h and wrote MTX_LOCK_SPIN
and MTX_UNLOCK_SPIN asm macros for the i386/mutex.h as we presently
need those.
Finally, caught up to the interface changes in all sys code.
Contributors: jake, jhb, jasone (in no particular order)
the index of the pollfd array to the number of fd's currently open, not
the maximum number of fd's. ie: if you had 0,1,2 open, you could not
use pollfd slots higher than 20. The specs say we only have to support
OPEN_MAX [64] entries but we allow way more than that.
Deal with excessive dirty buffers when msync() syncs non-contiguous
dirty buffers by checking for the case in UFS *before* checking for
clusterability.
Pre-rfork code assumed inherent locking of a process's file descriptor
array. However, with the advent of rfork() the file descriptor table
could be shared between processes. This patch closes over a dozen
serious race conditions related to one thread manipulating the table
(e.g. closing or dup()ing a descriptor) while another is blocked in
an open(), close(), fcntl(), read(), write(), etc...
PR: kern/11629
Discussed with: Alexander Viro <viro@math.psu.edu>
"kern/sys_generic.c:358: warning: cast discards qualifiers from pointer
target type"
The idea for using the uintptr_t intermediate cast for de-constifying
a pointer was hinted at by bde some time ago.
operation or after it. If the ktrace operation was enabled while the
process was blocked doing IO, the race would allow it to pass down
invalid (uninitialized) data and panic later down the call stack.
never expire if poll() or select() was called before the system had been
in multiuser for 1 second. This was caused by only checking to see if
tv_sec was zero rather than checking both tv_sec and tv_usec.
instead of a struct iovec * array and int len. Get rid of stupidly trying
to allocate all of the memory and copyin()ing the entire iovec[], and
instead just do the proper VOP_WRITE() in ktrwrite() using a copy of
the struct uio that the syscall originally used.
This solves the DoS which could easily be performed; to work around the
DoS, one could also remove "options KTRACE" from the kernel. This is
a very strong MFC candidate for 4.1.
Found by: art@OpenBSD.org
fd's in the range of 32-63, 96-127 etc. The first problem was the
FD_*() macros were shifting a 32 bit integer "1" left by more than
32 bits. The same problem happened in selscan(). ffs() also takes
an int argument and causes failure. For cases where int == long
(ie: the usual case for x86, but not always as gcc can have long
being a 64 bit quantity) ffs() could be used.
Reported by: Marian Stagarescu <marian@bile.skycache.com>
Reviewed by: dfr, gallatin (sys/types.h only)
Approved by: jkh
Make gratuitous style(9) fixes (me, not the submitter) to make the aio
code more readable.
PR: kern/12053
Submitted by: Chris Sedore <cmsedore@maxwell.syr.edu>
far-reaching in fd-land, so you'll want to consult the code for
changes. The biggest change is that now, you don't use
fp->f_ops->fo_foo(fp, bar)
but instead
fo_foo(fp, bar),
which increments and decrements the fp refcount upon entry and exit.
Two new calls, fhold() and fdrop(), are provided. Each does what it
seems like it should, and if fdrop() brings the refcount to zero, the
fd is freed as well.
Thanks to peter ("to hell with it, it looks ok to me.") for his review.
Thanks to msmith for keeping me from putting locks everywhere :)
Reviewed by: peter
NetBSD compatible.
Add parameter to fo_read and fo_write. (The only flag FOF_OFFSET mean that
the offset is set in the struct uio).
Factor out some common code from read/pread/write/pwrite syscalls.
c_caddr_t with extreme prejudice. Here the point of the original
cast to caddr_t was to break the warning about the const mismatch
between write(2)'s `const void *buf' and `struct uio's `char
*iov_base' (previous bitrot gave a gratuitous dependency on caddr_t
being char *). Compiling with -Wcast-qual made the cast a full
no-op.
This change has no effect on the warning for discarding `const'
on assignment to iov_base. The warning should not be fixed by
splitting `struct iovec' into a non-const version for read()
and a const version for write(), since correct const poisoning
would affect all pointers to i/o addresses. Const'ness should
probably be forgotten by not declaring it in syscalls.master.
0. This makes it difficult to do efficient manipulation of the
struct pollfd since you can't leave a slot empty.
PR: 8599
Submitted-by: Marc Slemko <marcs@znep.com>
by bde, a few other tweaks to get the patch to apply cleanly again and
some improvements to the comments.
This change closes some fairly minor security holes associated with
F_SETOWN, fixes a few bugs, and removes some limitations that F_SETOWN
had on tty devices. For more details, see the description on the PR.
Because this patch increases the size of the proc and pgrp structures,
it is necessary to re-install the includes and recompile libkvm,
the vinum lkm, fstat, gcore, gdb, ipfilter, ps, top, and w.
PR: kern/7899
Reviewed by: bde, elvind
Add some overflow checks to read/write (from bde).
Change all modifications to vm_page::flags, vm_page::busy, vm_object::flags
and vm_object::paging_in_progress to use operations which are not
interruptable.
Reviewed by: Bruce Evans <bde@zeta.org.au>
* Figure out UTC relative to boottime. Four new functions provide
time relative to boottime.
* move "runtime" into struct proc. This helps fix the calcru()
problem in SMP.
* kill mono_time.
* add timespec{add|sub|cmp} macros to time.h. (XXX: These may change!)
* nanosleep, select & poll takes long sleeps one day at a time
Reviewed by: bde
Tested by: ache and others
"time" wasn't a atomic variable, so splfoo() protection were needed
around any access to it, unless you just wanted the seconds part.
Most uses of time.tv_sec now uses the new variable time_second instead.
gettime() changed to getmicrotime(0.
Remove a couple of unneeded splfoo() protections, the new getmicrotime()
is atomic, (until Bruce sets a breakpoint in it).
A couple of places needed random data, so use read_random() instead
of mucking about with time which isn't random.
Add a new nfs_curusec() function.
Mark a couple of bogosities involving the now disappeard time variable.
Update ffs_update() to avoid the weird "== &time" checks, by fixing the
one remaining call that passwd &time as args.
Change profiling in ncr.c to use ticks instead of time. Resolution is
the same.
Add new function "tvtohz()" to avoid the bogus "splfoo(), add time, call
hzto() which subtracts time" sequences.
Reviewed by: bde
it in struct proc instead.
This fixes a boatload of compiler warning, and removes a lot of cruft
from the sources.
I have not removed the /*ARGSUSED*/, they will require some looking at.
libkvm, ps and other userland struct proc frobbing programs will need
recompiled.
Distribute all but the most fundamental malloc types. This time I also
remembered the trick to making things static: Put "static" in front of
them.
A couple of finer points by: bde
<sys/ttycom.h> and sometimes <sys/filio.h> instead of <sys/ioctl.h>
in miscellaneous files. Most of these files have nothing to do
with ttys but need to include <sys/ttycom.h> to get the definitions
of TIOC[SG]PGRP which are (ab)used to convert F[SG]ETOWN fcntls into
ioctls.
null casts. `time' is nonvolatile for accesses within a region locked
by splclock()/splx(). Accesses outside such a region are invalid, and
splx() must have the side effect of potentially changing all global
variables (since there are hundreds of sort of volatile variables like
`time'), so declaring `time' as volatile didn't have any real benefits.
- avoid malloc() if the number of fds is small.
- pack the bits better so that `small' is quite large.
- don't waste time generating zero bits for null fd_set pointers or
scanning these bits.
Possibly improved select():
- free malloc()ed storage before returning. This is simpler and I
think huge select()s aren't worth optimizing since they are rare,
relative gain would be small and there would be tiny costs for all
selects().
Reviewed by: ache (first version by him too)
This will make a number of things easier in the future, as well as (finally!)
avoiding the Id-smashing problem which has plagued developers for so long.
Boy, I'm glad we're not using sup anymore. This update would have been
insane otherwise.
structs and prototypes for syscalls.
Ifdefed duplicated decentralized declarations of args structs. It's
convenient to have this visible but they are hard to maintain. Some
are already different from the central declarations. 4.4lite2 puts
them in comments in the function headers but I wanted to avoid the
large changes for that.
cycles. While waiting there I added a lot of the extra ()'s I have, (I have
never used LISP to any extent). So I compiled the kernel with -Wall and
shut up a lot of "suggest you add ()'s", removed a bunch of unused var's
and added a couple of declarations here and there. Having a lap-top is
highly recommended. My kernel still runs, yell at me if you kernel breaks.
