1. Printing large quads in small bases overflowed the buffer if
sizeof(u_quad_t) > sizeof(u_long).
2. The sharpflag checks had operator precedence bugs due to excessive
parentheses in all the wrong places.
3. The explicit 0L was bogus in the quad_t comparison and useless in
the long comparision.
4. There was some more bitrot in the comment about ksprintn(). Our
ksprintn() handles bases up to 36 as well as down to 2.
Bruce has other complaints about using %q in kernel and would rather
we went towards using the C9X style %ll and/or %j. (I agree for that
matter, as long as gcc/egcs know how to deal with that.)
Submitted by: bde
I don't know if it was intentional or not, but it would have printed out:
/compat/linux/foo/bar.so: interpreter not found
If it was, then I've broken it. De-constifying the 'interp' variable
or carrying the constness through to elf_load_file() are alternatives.
Alpha believes that %q is for long long, whereas our quad_t and int64_t
is only just a plain long. long long on the alpha is the same size (64
bit) as a long. It was requested, but I have not implemented yet, support
for C9X style %lld - it should be pretty easy though.
large (1G) memory machine configurations. I was able to run 'dbench 32'
on a 32MB system without bring the machine to a grinding halt.
* buffer cache hash table now dynamically allocated. This will
have no effect on memory consumption for smaller systems and
will help scale the buffer cache for larger systems.
* minor enhancement to pmap_clearbit(). I noticed that
all the calls to it used constant arguments. Making
it an inline allows the constants to propogate to
deeper inlines and should produce better code.
* removal of inherent vfs_ioopt support through the emplacement
of appropriate #ifdef's, with John's permission. If we do not
find a use for it by the end of the year we will remove it entirely.
* removal of getnewbufloops* counters & sysctl's - no longer
necessary for debugging, getnewbuf() is now optimal.
* buffer hash table functions removed from sys/buf.h and localized
to vfs_bio.c
* VFS_BIO_NEED_DIRTYFLUSH flag and support code added
( bwillwrite() ), allowing processes to block when too many dirty
buffers are present in the system.
* removal of a softdep test in bdwrite() that is no longer necessary
now that bdwrite() no longer attempts to flush dirty buffers.
* slight optimization added to bqrelse() - there is no reason
to test for available buffer space on B_DELWRI buffers.
* addition of reverse-scanning code to vfs_bio_awrite().
vfs_bio_awrite() will attempt to locate clusterable areas
in both the forward and reverse direction relative to the
offset of the buffer passed to it. This will probably not
make much of a difference now, but I believe we will start
to rely on it heavily in the future if we decide to shift
some of the burden of the clustering closer to the actual
I/O initiation.
* Removal of the newbufcnt and lastnewbuf counters that Kirk
added. They do not fix any race conditions that haven't already
been fixed by the gbincore() test done after the only call
to getnewbuf(). getnewbuf() is a static, so there is no chance
of it being misused by other modules. ( Unless Kirk can think
of a specific thing that this code fixes. I went through it
very carefully and didn't see anything ).
* removal of VOP_ISLOCKED() check in flushbufqueues(). I do not
think this check is necessary, the buffer should flush properly
whether the vnode is locked or not. ( yes? ).
* removal of extra arguments passed to getnewbuf() that are not
necessary.
* missed cluster_wbuild() that had to be a cluster_wbuild_wb() in
vfs_cluster.c
* vn_write() now calls bwillwrite() *PRIOR* to locking the vnode,
which should greatly aid flushing operations in heavy load
situations - both the pageout and update daemons will be able
to operate more efficiently.
* removal of b_usecount. We may add it back in later but for now
it is useless. Prior implementations of the buffer cache never
had enough buffers for it to be useful, and current implementations
which make more buffers available might not benefit relative to
the amount of sophistication required to implement a b_usecount.
Straight LRU should work just as well, especially when most things
are VMIO backed. I expect that (even though John will not like
this assumption) directories will become VMIO backed some point soon.
Submitted by: Matthew Dillon <dillon@backplane.com>
Reviewed by: Kirk McKusick <mckusick@mckusick.com>
than a review, this was a nice puzzle.
This is supposed to be binary and source compatible with older
applications that access the old FreeBSD-style three arguments to a
signal handler.
Except those applications that access hidden signal handler arguments
bejond the documented third one. If you have applications that do,
please let me know so that we take the opportunity to provide the
functionality they need in a documented manner.
Also except application that use 'struct sigframe' directly. You need
to recompile gdb and doscmd. `make world` is recommended.
Example program that demonstrates how SA_SIGINFO and old-style FreeBSD
handlers (with their three args) may be used in the same process is at
http://www3.cons.org/tmp/fbsd-siginfo.c
Programs that use the old FreeBSD-style three arguments are easy to
change to SA_SIGINFO (although they don't need to, since the old style
will still work):
Old args to signal handler:
void handler_sn(int sig, int code, struct sigcontext *scp)
New args:
void handler_si(int sig, siginfo_t *si, void *third)
where:
old:code == new:second->si_code
old:scp == &(new:si->si_scp) /* Passed by value! */
The latter is also pointed to by new:third, but accessing via
si->si_scp is preferred because it is type-save.
FreeBSD implementation notes:
- This is just the framework to make the interface POSIX compatible.
For now, no additional functionality is provided. This is supposed
to happen now, starting with floating point values.
- We don't use 'sigcontext_t.si_value' for now (POSIX meant it for
realtime-related values).
- Documentation will be updated when new functionality is added and
the exact arguments passed are determined. The comments in
sys/signal.h are meant to be useful.
Reviewed by: BDE
dynamicly linked binaries to run in a chroot'd environment with "emul_path"
as the new root. The new behavior of loading interpreters is identical to the
principle of overlaying.
PR: 10145
into uipc_mbuf.c. This reduces three sets of identical tunable code to
one set, and puts the initialisation with the mbuf code proper.
Make NMBUFs tunable as well.
Move the nmbclusters sysctl here as well.
Move the initialisation of maxsockets from param.c to uipc_socket2.c,
next to its corresponding sysctl.
Use the new tunable macros for the kern.vm.kmem.size tunable (this should have
been in a separate commit, whoops).
returns 0 after ptrace() attach and/or detach doesn't quite quite
deliver a signal. Perhaps the process shouldn't be woken in this
case, but avoiding the problem is easy.
PR: 12247
Fixed a couple of places where mechanical fixing of compiler warnings
caused misspelling of NOLOCKF as NULL.
allow changes to the filesystem's write_behind behavior. By the
default the filesystem aggressively issues write_behind's. Three values
may be specified for vfs.write_behind. 0 disables write_behind, 1 results
in historical operation (agressive write_behind), and 2 is an experimental
backed-off write_behind. The values of 0 and 1 are recommended. The value
of 0 is recommended in conjuction with an increase in the number of
NBUF's and the number of dirty buffers allowed (vfs.{lo,hi}dirtybuffers).
Note that a value of 0 will radically increase the dirty buffer load on
the system. Future work on write_behind behavior will use values 2 and
greater for testing purposes.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
Reviewed by: Kirk McKusick <mckusick@mckusick.com>
QUEUE_AGE, QUEUE_LRU, and QUEUE_EMPTY we instead have QUEUE_CLEAN,
QUEUE_DIRTY, QUEUE_EMPTY, and QUEUE_EMPTYKVA. With this patch clean
and dirty buffers have been separated. Empty buffers with KVM
assignments have been separated from truely empty buffers. getnewbuf()
has been rewritten and now operates in a 100% optimal fashion. That is,
it is able to find precisely the right kind of buffer it needs to
allocate a new buffer, defragment KVM, or to free-up an existing buffer
when the buffer cache is full (which is a steady-state situation for
the buffer cache).
Buffer flushing has been reorganized. Previously buffers were flushed
in the context of whatever process hit the conditions forcing buffer
flushing to occur. This resulted in processes blocking on conditions
unrelated to what they were doing. This also resulted in inappropriate
VFS stacking chains due to multiple processes getting stuck trying to
flush dirty buffers or due to a single process getting into a situation
where it might attempt to flush buffers recursively - a situation that
was only partially fixed in prior commits. We have added a new daemon
called the buf_daemon which is responsible for flushing dirty buffers
when the number of dirty buffers exceeds the vfs.hidirtybuffers limit.
This daemon attempts to dynamically adjust the rate at which dirty buffers
are flushed such that getnewbuf() calls (almost) never block.
The number of nbufs and amount of buffer space is now scaled past the
8MB limit that was previously imposed for systems with over 64MB of
memory, and the vfs.{lo,hi}dirtybuffers limits have been relaxed
somewhat. The number of physical buffers has been increased with the
intention that we will manage physical I/O differently in the future.
reassignbuf previously attempted to keep the dirtyblkhd list sorted which
could result in non-deterministic operation under certain conditions,
such as when a large number of dirty buffers are being managed. This
algorithm has been changed. reassignbuf now keeps buffers locally sorted
if it can do so cheaply, and otherwise gives up and adds buffers to
the head of the dirtyblkhd list. The new algorithm is deterministic but
not perfect. The new algorithm greatly reduces problems that previously
occured when write_behind was turned off in the system.
The P_FLSINPROG proc->p_flag bit has been replaced by the more descriptive
P_BUFEXHAUST bit. This bit allows processes working with filesystem
buffers to use available emergency reserves. Normal processes do not set
this bit and are not allowed to dig into emergency reserves. The purpose
of this bit is to avoid low-memory deadlocks.
A small race condition was fixed in getpbuf() in vm/vm_pager.c.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
Reviewed by: Kirk McKusick <mckusick@mckusick.com>
SYSINIT_KT() etc (which is a static, compile-time procedure), use a
NetBSD-style kthread_create() interface. kproc_start is still available
as a SYSINIT() hook. This allowed simplification of chunks of the
sysinit code in the process. This kthread_create() is our old kproc_start
internals, with the SYSINIT_KT fork hooks grafted in and tweaked to work
the same as the NetBSD one.
One thing I'd like to do shortly is get rid of nfsiod as a user initiated
process. It makes sense for the nfs client code to create them on the
fly as needed up to a user settable limit. This means that nfsiod
doesn't need to be in /sbin and is always "available". This is a fair bit
easier to do outside of the SYSINIT_KT() framework.
the caller can easily find the child proc struct. fork(), rfork() etc
syscalls set p->p_retval[] themselves. Simplify the SYSINIT_KT() code
and other kernel thread creators to not need to use pfind() to find the
child based on the pid. While here, partly tidy up some of the fork1()
code for RF_SIGSHARE etc.
(really this time) fix pageout to swap and a couple of clustering cases.
This simplifies BUF_KERNPROC() so that it unconditionally reassigns the
lock owner rather than testing B_ASYNC and having the caller decide when
to do the reassign. At present this is required because some places use
B_CALL/b_iodone to free the buffers without B_ASYNC being set. Also,
vfs_cluster.c explicitly calls BUF_KERNPROC() when attaching the buffers
rather than the parent walking the cluster_head tailq.
Reviewed by: Kirk McKusick <mckusick@mckusick.com>
if LK_RECURSIVE is not set, as we will simply return that the
lock is busy and not actually deadlock. This allows processes
to use polling locks against buffers that they may already
hold exclusively locked.
lockmgr locks. This commit should be functionally equivalent to the old
semantics. That is, all buffer locking is done with LK_EXCLUSIVE
requests. Changes to take advantage of LK_SHARED and LK_RECURSIVE will
be done in future commits.
- Split syscons source code into manageable chunks and reorganize
some of complicated functions.
- Many static variables are moved to the softc structure.
- Added a new key function, PREV. When this key is pressed, the vty
immediately before the current vty will become foreground. Analogue
to PREV, which is usually assigned to the PrntScrn key.
PR: kern/10113
Submitted by: Christian Weisgerber <naddy@mips.rhein-neckar.de>
- Modified the kernel console input function sccngetc() so that it
handles function keys properly.
- Reorganized the screen update routine.
- VT switching code is reorganized. It now should be slightly more
robust than before.
- Added the DEVICE_RESUME function so that syscons no longer hooks the
APM resume event directly.
- New kernel configuration options: SC_NO_CUTPASTE, SC_NO_FONT_LOADING,
SC_NO_HISTORY and SC_NO_SYSMOUSE.
Various parts of syscons can be omitted so that the kernel size is
reduced.
SC_PIXEL_MODE
Made the VESA 800x600 mode an option, rather than a standard part of
syscons.
SC_DISABLE_DDBKEY
Disables the `debug' key combination.
SC_ALT_MOUSE_IMAGE
Inverse the character cell at the mouse cursor position in the text
console, rather than drawing an arrow on the screen.
Submitted by: Nick Hibma (n_hibma@FreeBSD.ORG)
SC_DFLT_FONT
makeoptions "SC_DFLT_FONT=_font_name_"
Include the named font as the default font of syscons. 16-line,
14-line and 8-line font data will be compiled in. This option replaces
the existing STD8X16FONT option, which loads 16-line font data only.
- The VGA driver is split into /sys/dev/fb/vga.c and /sys/isa/vga_isa.c.
- The video driver provides a set of ioctl commands to manipulate the
frame buffer.
- New kernel configuration option: VGA_WIDTH90
Enables 90 column modes: 90x25, 90x30, 90x43, 90x50, 90x60. These
modes are mot always supported by the video card.
PR: i386/7510
Submitted by: kbyanc@freedomnet.com and alexv@sui.gda.itesm.mx.
- The header file machine/console.h is reorganized; its contents is now
split into sys/fbio.h, sys/kbio.h (a new file) and sys/consio.h
(another new file). machine/console.h is still maintained for
compatibility reasons.
- Kernel console selection/installation routines are fixed and
slightly rebumped so that it should now be possible to switch between
the interanl kernel console (sc or vt) and a remote kernel console
(sio) again, as it was in 2.x, 3.0 and 3.1.
- Screen savers and splash screen decoders
Because of the header file reorganization described above, screen
savers and splash screen decoders are slightly modified. After this
update, /sys/modules/syscons/saver.h is no longer necessary and is
removed.
at which we may sleep. So, after completing our buffer allocation
we must ensure that another process has not come along and allocated
a different buffer with the same identity. We do this by keeping a
global counter of the number of buffers that getnewbuf has allocated.
We save this count when we enter getnewbuf and check it when we are
about to return. If it has changed, then other buffers were allocated
while we were in getnewbuf, so we must return NULL to let our parent
know that it must recheck to see if it still needs the new buffer.
Hopefully this fix will eliminate the creation of duplicate buffers
with the same identity and the obscure corruptions that they cause.
done is less-than cute, but this whole file is suffering from some amount
of bitrot. Execution of zipped files should probably be implemented in a
manner similar to that of #!/interpreted files.
PR: kern/10780
automatically hacks on the active copy of the IDT if f00f_hack()
has changed it. This also allows simplifications in setidt().
This fixes breakage of FP exception handling by rev.1.55 of
sys/kernel.h. FP exceptions were sent to npx.c's probe handlers
because npx.c "restored" the old handlers to the wrong copy of the
IDT. The SYSINIT for f00f_hack() was purposely run quite late to
avoid problems like this, but it is bogusly associated with the
SYSINIT for proc0 so it was moved with the latter.
Problem reported and fix tested by: Martin Cracauer <cracauer@cons.org>
This is the change to struct sockets that gets rid of so_uid and replaces
it with a much more useful struct pcred *so_cred. This is here to be able
to do socket-level credential checks (i.e. IPFW uid/gid support, to be added
to HEAD soon). Along with this comes an update to pidentd which greatly
simplifies the code necessary to get a uid from a socket. Soon to come:
a sysctl() interface to finding individual sockets' credentials.
would occur when clustering them - caused by running out of buffers
and taking a degenerate code path as a result. It appears that waiting
instead for buffers to become available is okay.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
Discovered by: Craig A Soules <soules+@andrew.cmu.edu>
- this causes POSIX locking to use the thread group leader
(p->p_leader) as the locking thread for all advisory locks.
In non-kernel-threaded code p->p_leader == p, so this will have
no effect.
This results in (more) correct POSIX threaded flock-ing semantics.
It also prevents the leader from exiting before any of the children.
(so that p->p_leader will never be stale) in exit1().
We have been running this patch for over a month now in our lab
under load and at customer sites.
Submitted by: John Plevyak <jplevyak@inktomi.com>
to either enqueue or free their mbuf chains, but tcp_usr_send() was
dropping them on the floor if the tcpcb/inpcb has been torn down in the
middle of a send/write attempt. This has been responsible for a wide
variety of mbuf leak patterns, ranging from slow gradual leakage to rather
rapid exhaustion. This has been a problem since before 2.2 was branched
and appears to have been fixed in rev 1.16 and lost in 1.23/1.28.
Thanks to Jayanth Vijayaraghavan <jayanth@yahoo-inc.com> for checking
(extensively) into this on a live production 2.2.x system and that it
was the actual cause of the leak and looks like it fixes it. The machine
in question was loosing (from memory) about 150 mbufs per hour under
load and a change similar to this stopped it. (Don't blame Jayanth
for this patch though)
An alternative approach to this would be to recheck SS_CANTSENDMORE etc
inside the splnet() right before calling pru_send() after all the potential
sleeps, interrupts and delays have happened. However, this would mean
exposing knowledge of the tcp stack's reset handling and removal of the
pcb to the generic code. There are other things that call pru_send()
directly though.
Problem originally noted by: John Plevyak <jplevyak@inktomi.com>
The cdevsw_add() function now finds the major number(s) in the
struct cdevsw passed to it. cdevsw_add_generic() is no longer
needed, cdevsw_add() does the same thing.
cdevsw_add() will print an message if the d_maj field looks bogus.
Remove nblkdev and nchrdev variables. Most places they were used
bogusly. Instead check a dev_t for validity by seeing if devsw()
or bdevsw() returns NULL.
Move bdevsw() and devsw() functions to kern/kern_conf.c
Bump __FreeBSD_version to 400006
This commit removes:
72 bogus makedev() calls
26 bogus SYSINIT functions
if_xe.c bogusly accessed cdevsw[], author/maintainer please fix.
I4b and vinum not changed. Patches emailed to authors. LINT
probably broken until they catch up.
Reformat and initialize correctly all "struct cdevsw".
Initialize the d_maj and d_bmaj fields.
The d_reset field was not removed, although it is never used.
I used a program to do most of this, so all the files now use the
same consistent format. Please keep it that way.
Vinum and i4b not modified, patches emailed to respective authors.
easier to use and more flexible.
* Change BUS_ADD_CHILD to take an order argument instead of a place.
* Define a partial ordering for isa devices so that sensitive devices are
probed before non-sensitive ones.
if there is no character device associated with the block device. In this
case that doesn't matter because bdevvp() doesn't use the character
device structure.
I can use the pointy bit of the axe too.
It never makes sense to specify MAP_COPY_NEEDED without also specifying
MAP_COPY_ON_WRITE, and vice versa. Thus, MAP_COPY_ON_WRITE suffices.
Reviewed by: David Greenman <dg@root.com>
instances to a parent bus.
* Define a new method BUS_ADD_CHILD which can be called from DEVICE_IDENTIFY
to add new instances.
* Add a generic implementation of DEVICE_PROBE which calls DEVICE_IDENTIFY
for each driver attached to the parent's devclass.
* Move the hint-based isa probe from the isa driver to a new isahint driver
which can be shared between i386 and alpha.
inodes were synced every 15 seconds. This is now reversed as during
directory create, we cannot commit the directory entry until its
inode has been written. With this switch, the inodes will be more
likely to be written by the time that the directory is written thus
reducing the number of directory rollbacks that are needed.
Changed to `const void *'. utrace() is undocumented, so nothing should
notice.
Fixed missing consts for utrace() and ktrace() in syscalls.master.
sys/ktrace.h is missing some Lite2 changes of shorts to ints.
with malloc type at the tail of the list changed the list from
linear to circular. This seemed to cause surprisingly few problems,
but it now causes weird output from `vmstat -m', probably because
a more important malloc type is now at the tail of the list.
Fix it by abusing ks_limit instead of ks_next as a flag for being
on the list. Don't forget to clear the flag when a malloc type is
uninit'ed. Uninit'ing is still fundamentally broken -- it loses
history.
udev_t in the kernel but still called dev_t in userland.
Provide functions to manipulate both types:
major() umajor()
minor() uminor()
makedev() umakedev()
dev2udev() udev2dev()
For now they're functions, they will become in-line functions
after one of the next two steps in this process.
Return major/minor/makedev to macro-hood for userland.
Register a name in cdevsw[] for the "filedescriptor" driver.
In the kernel the udev_t appears in places where we have the
major/minor number combination, (ie: a potential device: we
may not have the driver nor the device), like in inodes, vattr,
cdevsw registration and so on, whereas the dev_t appears where
we carry around a reference to a actual device.
In the future the cdevsw and the aliased-from vnode will be hung
directly from the dev_t, along with up to two softc pointers for
the device driver and a few houskeeping bits. This will essentially
replace the current "alias" check code (same buck, bigger bang).
A little stunt has been provided to try to catch places where the
wrong type is being used (dev_t vs udev_t), if you see something
not working, #undef DEVT_FASCIST in kern/kern_conf.c and see if
it makes a difference. If it does, please try to track it down
(many hands make light work) or at least try to reproduce it
as simply as possible, and describe how to do that.
Without DEVT_FASCIST I belive this patch is a no-op.
Stylistic/posixoid comments about the userland view of the <sys/*.h>
files welcome now, from userland they now contain the end result.
Next planned step: make all dev_t's refer to the same devsw[] which
means convert BLK's to CHR's at the perimeter of the vnodes and
other places where they enter the game (bootdev, mknod, sysctl).
unp_internalize() takes a reference to the descriptor. If the send
fails after unp_internalize(), the control mbuf would be freed ophaning
the reference.
Tested in -CURRENT by: Pierre Beyssac <beyssac@enst.fr>
through' to the C compiler.
* Allow the interface to specify a default implementation for methods.
* Allow 'static' methods which are not device specific.
* Add a simple scheme for probe routines to return a priority value. To
make life simple, priority values are negative numbers (positive numbers
are standard errno codes) with zero being the highest priority. The
driver which returns the highest priority will be chosen for the device.
upset about it (and generate things like __main() calls that are reserved
for main()). Renaming was phk's suggestion, but I'd already thought about
it too. (phk liked my suggested name tada() but I decided against it :-)
Reviewed by: phk
- first program lock a region in a file,
- second program wait on the lock,
- first program extend the region,
- second program interrupted by a signal.
particularly annoying hack, namely having the linker bash the moduledata
to set the container pointer, preventing it being const. In the process,
a stack of warnings were fixed and will probably allow a revisit of the
const C_SYSINIT() changes. This explicitly registers modules in files or
preload areas with the module system first, and let them initialize via
SYSINIT/DECLARE_MODULE later in their SI_ORDER_xxx order. The kludge of
finding the containing file is no longer needed since the registration
of modules onto the modules list is done in the context of initializing
the linker file.
Made a new (inline) function devsw(dev_t dev) and substituted it.
Changed to the BDEV variant to this format as well: bdevsw(dev_t dev)
DEVFS will eventually benefit from this change too.
Virtualize bdevsw[] from cdevsw. bdevsw() is now an (inline)
function.
Join CDEV_MODULE and BDEV_MODULE to DEV_MODULE (please pay attention
to the order of the cmaj/bmaj arguments!)
Join CDEV_DRIVER_MODULE and BDEV_DRIVER_MODULE to DEV_DRIVER_MODULE
(ditto!)
(Next step will be to convert all bdev dev_t's to cdev dev_t's
before they get to do any damage^H^H^H^H^H^Hwork in the kernel.)
for elf kernels (it is broken for all kernels due to lack of egcs support).
Renaming of many assembler labels is avoided by declaring by declaring
the labels that need to be visible to gprof as having type "function"
and depending on the elf version of gprof being zealous about discarding
the others. A few type declarations are still missing, mainly for SMP.
PR: 9413
Submitted by: Assar Westerlund <assar@sics.se> (initial parts)
INIT_PATH config option.
Also fix two bugs which caused an infinite loop in none of the programs
in the init_path were found. That code was obviously not tested!
piecemeal, middle-of-file writes for NFS. These hacks have caused no
end of trouble, especially when combined with mmap(). I've removed
them. Instead, NFS will issue a read-before-write to fully
instantiate the struct buf containing the write. NFS does, however,
optimize piecemeal appends to files. For most common file operations,
you will not notice the difference. The sole remaining fragment in
the VFS/BIO system is b_dirtyoff/end, which NFS uses to avoid cache
coherency issues with read-merge-write style operations. NFS also
optimizes the write-covers-entire-buffer case by avoiding the
read-before-write. There is quite a bit of room for further
optimization in these areas.
The VM system marks pages fully-valid (AKA vm_page_t->valid =
VM_PAGE_BITS_ALL) in several places, most noteably in vm_fault. This
is not correct operation. The vm_pager_get_pages() code is now
responsible for marking VM pages all-valid. A number of VM helper
routines have been added to aid in zeroing-out the invalid portions of
a VM page prior to the page being marked all-valid. This operation is
necessary to properly support mmap(). The zeroing occurs most often
when dealing with file-EOF situations. Several bugs have been fixed
in the NFS subsystem, including bits handling file and directory EOF
situations and buf->b_flags consistancy issues relating to clearing
B_ERROR & B_INVAL, and handling B_DONE.
getblk() and allocbuf() have been rewritten. B_CACHE operation is now
formally defined in comments and more straightforward in
implementation. B_CACHE for VMIO buffers is based on the validity of
the backing store. B_CACHE for non-VMIO buffers is based simply on
whether the buffer is B_INVAL or not (B_CACHE set if B_INVAL clear,
and vise-versa). biodone() is now responsible for setting B_CACHE
when a successful read completes. B_CACHE is also set when a bdwrite()
is initiated and when a bwrite() is initiated. VFS VOP_BWRITE
routines (there are only two - nfs_bwrite() and bwrite()) are now
expected to set B_CACHE. This means that bowrite() and bawrite() also
set B_CACHE indirectly.
There are a number of places in the code which were previously using
buf->b_bufsize (which is DEV_BSIZE aligned) when they should have
been using buf->b_bcount. These have been fixed. getblk() now clears
B_DONE on return because the rest of the system is so bad about
dealing with B_DONE.
Major fixes to NFS/TCP have been made. A server-side bug could cause
requests to be lost by the server due to nfs_realign() overwriting
other rpc's in the same TCP mbuf chain. The server's kernel must be
recompiled to get the benefit of the fixes.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
"use" and replaces it with equivalent inline code. The reason is that
Perl has some very nasty circular dependancies, and I am trying to
get the System Perl upgraded by one maintenance level.
The basic rule, until I can find a way to solve this, is that
the build tools MAY NOT use any library code; it must all be inline.
range attributes after they have been extracted from the master.
Hook up the i686 MP code to do this for each AP.
Be more careful about printing the default memory type for the i686.
Suggestions from: luoqi
This patch also moves the bogus comment (the comment is still not quite
right) and (as a side effect) removes some verbose initialisations (we
depend on static initialisation to 0 for almost everything in proc0).
The alpha kernels are bootable again. The change won't affect i386's
until machdep.c is changed.
Submitted by: bde
In heavy-writing situations, QUEUE_LRU can contain a large number
of DELWRI buffers at its head. These buffers must be moved
to the tail if they cannot be written async in order to reduce
the scanning time required to skip past these buffers in later
getnewbuf() calls.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
This is a seriously beefed up chroot kind of thing. The process
is jailed along the same lines as a chroot does it, but with
additional tough restrictions imposed on what the superuser can do.
For all I know, it is safe to hand over the root bit inside a
prison to the customer living in that prison, this is what
it was developed for in fact: "real virtual servers".
Each prison has an ip number associated with it, which all IP
communications will be coerced to use and each prison has its own
hostname.
Needless to say, you need more RAM this way, but the advantage is
that each customer can run their own particular version of apache
and not stomp on the toes of their neighbors.
It generally does what one would expect, but setting up a jail
still takes a little knowledge.
A few notes:
I have no scripts for setting up a jail, don't ask me for them.
The IP number should be an alias on one of the interfaces.
mount a /proc in each jail, it will make ps more useable.
/proc/<pid>/status tells the hostname of the prison for
jailed processes.
Quotas are only sensible if you have a mountpoint per prison.
There are no privisions for stopping resource-hogging.
Some "#ifdef INET" and similar may be missing (send patches!)
If somebody wants to take it from here and develop it into
more of a "virtual machine" they should be most welcome!
Tools, comments, patches & documentation most welcome.
Have fun...
Sponsored by: http://www.rndassociates.com/
Run for almost a year by: http://www.servetheweb.com/
- %fs register is added to trapframe and saved/restored upon kernel entry/exit.
- Per-cpu pages are no longer mapped at the same virtual address.
- Each cpu now has a separate gdt selector table. A new segment selector
is added to point to per-cpu pages, per-cpu global variables are now
accessed through this new selector (%fs). The selectors in gdt table are
rearranged for cache line optimization.
- fask_vfork is now on as default for both UP and SMP.
- Some aio code cleanup.
Reviewed by: Alan Cox <alc@cs.rice.edu>
John Dyson <dyson@iquest.net>
Julian Elischer <julian@whistel.com>
Bruce Evans <bde@zeta.org.au>
David Greenman <dg@root.com>
1:
s/suser/suser_xxx/
2:
Add new function: suser(struct proc *), prototyped in <sys/proc.h>.
3:
s/suser_xxx(\([a-zA-Z0-9_]*\)->p_ucred, \&\1->p_acflag)/suser(\1)/
The remaining suser_xxx() calls will be scrutinized and dealt with
later.
There may be some unneeded #include <sys/cred.h>, but they are left
as an exercise for Bruce.
More changes to the suser() API will come along with the "jail" code.
Get rid of the spl wrapper kludge, it doesn't seem to be needed between
init calls since all that's running is the domain/protocol timers and they
are safe since domain list modifications are splnet() protected (which
blocks the timers)
Interrupts under the new scheme are managed by the i386 nexus with the
awareness of the resource manager. There is further room for optimizing
the interfaces still. All the users of register_intr()/intr_create()
should be gone, with the exception of pcic and i386/isa/clock.c.
1. Make read-ahead work for pread and aio_read.
2. Fix one place where a comparison of uio_offset with -1
wasn't updated to use FOF_OFFSET.
3. Honor O_APPEND in the FOF_OFFSET case.
In addition, use the variable name "ioflag" in both vn_read and
vn_write to avoid possible confusion between the variable "flag"
and the parameter "flags".
Submitted by: Bruce Evans <bde@zeta.org.au> and me
i386 platform boots, it is no longer ISA-centric, and is fully dynamic.
Most old drivers compile and run without modification via 'compatability
shims' to enable a smoother transition. eisa, isapnp and pccard* are
not yet using the new resource manager. Once fully converted, all drivers
will be loadable, including PCI and ISA.
(Some other changes appear to have snuck in, including a port of Soren's
ATA driver to the Alpha. Soren, back this out if you need to.)
This is a checkpoint of work-in-progress, but is quite functional.
The bulk of the work was done over the last few years by Doug Rabson and
Garrett Wollman.
Approved by: core
for as much as one second, but no more. Allows a miscreant to
double-time march the clock, but no worse.
XXX Unlike putting negative deltas in a while(1), performing small
positive steps inside of a while(1) will return EPERM for the
unpermitted ones. Repeated negative deltas are clamped without
error (but the kernel does log a notice).
1 second prior to the highest the clock has run so far. This allows
time adjusters like xntpd to do their work, but the worst a miscreant
can do is "freeze" the clock, not go back in time.
We still need to decide on an algorithm to clamp positive adjustments.
As it stands, it is possible to achieve arbitrary negative adjustments
by "wrapping" time around.
PR: 10361
condition ( bufspace > hibufspace ), an inappropriate scan of the empty
queue was performed looking for buffer space to free up.
Submitted by: Matthew Dillon <dillon@apollo.backplane.com>
unallocated parts of the last page when the file ended on a frag
but not a page boundary.
Delimitted by tags PRE_MATT_MMAP_EOF and POST_MATT_MMAP_EOF,
in files alpha/alpha/pmap.c i386/i386/pmap.c nfs/nfs_bio.c vm/pmap.h
vm/vm_page.c vm/vm_page.h vm/vnode_pager.c miscfs/specfs/spec_vnops.c
ufs/ufs/ufs_readwrite.c kern/vfs_bio.c
Submitted by: Matt Dillon <dillon@freebsd.org>
Reviewed by: Alan Cox <alc@freebsd.org>
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.
the address of the ps_strings structure to the process via %ebx.
For other kinds of binaries, %ebx is still zeroed as before.
Submitted by: Thomas Stephens <tas@stephens.org>
Reviewed by: jdp
* bus_setup_intr() as a wrapper for BUS_SETUP_INTR
* bus_teardown_intr() as a wrapper for BUS_TEARDOWN_INTR
* device_get_nameunit() which returns e.g. "foo0" for name "foo" and unit 0.
* device_set_desc_copy() malloc a copy of the description string.
* device_quiet(), device_is_quiet(), device_verbose() suppress probe message.
Add one method to the BUS interface, BUS_CHILD_DETACHED() which is called
after the child has been detached to allow the bus to clean up any memory
which it allocated on behalf of the child.
I also fixed a bug which corrupted the list of drivers in a devclass if
a driver was added to more than one devclass.
if uio->uio_offset != -1. This fixes a problem with aio_read/write
and permits a straightforward implementation of pread/pwrite.
PR: kern/8669
Submitted by: John Plevyak <jplevyak@inktomi.com>
Reviewed by: Matthew Dillon <dillon@apollo.backplane.com>
kern.chroot_allow_open_directories = 0
chroot(2) fails if there are open directories.
kern.chroot_allow_open_directories = 1 (default)
chroot(2) fails if there are open directories and the process
is subject of a previous chroot(2).
kern.chroot_allow_open_directories = anything else
filedescriptors are not checked. (old behaviour).
I'm very interested in reports about software which breaks when
running with the default setting.
space. When doing this, it is possible to for another process to attempt
to get an exclusive lock on the vnode and deadlock the mmap/read
combination when the uiomove() call tries to obtain a second
shared lock on the vnode. There is still a potential deadlock
situation with write()/mmap().
Submitted by: Matt Dillon <dillon@freebsd.org>
Reviewed by: Luoqi Chen <luoqi@freebsd.org>
Delimmitted by tag PRE_MATT_MMAP_LOCK and POST_MATT_MMAP_LOCK
in kern/kern_lock.c kern/kern_subr.c
including alan, john, me, luoqi, and kirk
Submitted by: Matt Dillon <dillon@frebsd.org>
This change implements a relatively sophisticated fix to getnewbuf().
There were two problems with getnewbuf(). First, the writerecursion
can lead to a system stack overflow when you have NFS and/or VN
devices in the system. Second, the free/dirty buffer accounting was
completely broken. Not only did the nfs routines blow it trying to
manually account for the buffer state, but the accounting that was
done did not work well with the purpose of their existance: figuring
out when getnewbuf() needs to sleep.
The meat of the change is to kern/vfs_bio.c. The remaining diffs are
all minor except for NFS, which includes both the fixes for bp
interaction AND fixes for a 'biodone(): buffer already done' lockup.
Sys/buf.h also contains a chaining structure which is not used by
this patchset but is used by other patches that are coming soon.
This patch deliniated by tags PRE_MAT_GETBUF and POST_MAT_GETBUF.
(sorry for the missing T matt)
was discarded on every call to calcru(). Hacking on the `switchtime'
global for a related fix in rev.1.38 of kern_resource.c was too fragile
and broke when p_switchtime went away.
PR: 10402
This code is backwards compatible with the older "microkernel" PLL, but
allows ntpd v4 to use nanosecond resolution. Many other improvements.
PPS_SYNC and hardpps() are NOT supported yet.
unregister them after sysuninits when unloading.
* Add code to vfs_register() to set the oid number of vfs sysctls to
the type number of the filesystem.
Reviewed by: bde
to manage their own memory. Tested on my machine (make buildworld).
I've made analogous changes on the alpha, but don't have a machine
to test.
Not-objected-to by: dg, gibbs
numbers as chars or use bogus casts in an attempt to unmisrepresnt
them. In top, don't assume that 0xff is the only negative cpu
number when cpu numbers are (mis)represented.
Higher numbers led to smaller quanta.
In discussion with BDE, change this parameter to be in uSecs
to make it machine independent,
and limit it to non zero multiples of 'tick' (rounding down).
Also make the variabel globally available so that the present function that
returns its value (used for posix scheduling I believe) can go away.
Submitted by: Bruce Evans <bde@freebsd.org>
This produced races resulting in panics and filesystem corruptions
under some circumstances.
Reviewed by: luoqi chen <luoqi@freebsd.org>
Reviewed by: Kirk McKusick <mckusick@mckusick.com>
Submitted by: Matt Dillon <dillon@freebsd.org>
not per-process. Keep it in `switchtime' consistently.
It is now clear that the timestamp is always valid in fork_trampoline()
except when the child is running on a previously idle cpu, which
can only happen if there are multiple cpus, so don't check or set
the timestamp in fork_trampoline except in the (i386) SMP case.
Just remove the alpha code for setting it unconditionally, since
there is no SMP case for alpha and the code had rotted.
Parts reviewed by: dfr, phk
Add d_parms() to {c,b}devsw[]. If non-NULL this function points to
a device routine that will properly fill in the specinfo structure.
vfs_subr.c's checkalias() supplies appropriate defaults. This change
should be fully backwards compatible with existing devices.
often for it to be a good criterion for switching kernel cpu hogs --
it is true after most wakeups. Use the criterion "has been running
for >= 2 quanta" instead.
in "src/sys/sys/param.h".
Fix the ELF image activator so that it can handle dynamic linkers
which are executables linked at a fixed address. This improves
compliance with the ABI spec, and it opens the door to possibly
better dynamic linker performance in the future. I've experimented
a bit with a fixed-address dynamic linker, and it works fine. But
I don't have any measurements yet to determine whether it's
worthwhile.
Also, remove a few calculations that were never used for anything.
I will increment __FreeBSD_version, since this adds a new capability
to the kernel that the dynamic linker might some day rely upon.
being loaded twice. It used rindex() to strip the pathname but failed
to account for the fact that rindex() will return a pointer to the '/',
not the first character of the filename.
Submitted by: Nick Hibma <hibma@skylink.it>
is the preparation step for moving pmap storage out of vmspace proper.
Reviewed by: Alan Cox <alc@cs.rice.edu>
Matthew Dillion <dillon@apollo.backplane.com>
complaints about ps_refcnt greater than two when we try to fork() a
kthread from proc0 with RFSIGSHARE flag set.
Noticed by: Tor Egge <tegge@fast.no>
Reviewed by: Richard Seaman, Jr. <dick@tar.com>
This makes it possible to change the sysctl tree at runtime.
* Change KLD to find and register any sysctl nodes contained in the loaded
file and to unregister them when the file is unloaded.
Reviewed by: Archie Cobbs <archie@whistle.com>,
Peter Wemm <peter@netplex.com.au> (well they looked at it anyway)
peripheral drivers can determine where in the devstat(9) list they are
inserted.
This requires recompilation of libdevstat, systat, vmstat, rpc.rstatd, and
any ports that depend on the devstat code, since the size of the devstat
structure has changed. The devstat version number has been incremented as
well to reflect the change.
This sorts devices in the devstat list in "more interesting" to "less
interesting" order. So, for instance, da devices are now more important
than floppy drives, and so will appear before floppy drives in the default
output from systat, iostat, vmstat, etc.
The order of devices is, for now, kept in a central table in devicestat.h.
If individual drivers were able to make a meaningful decision on what
priority they should be at attach time, we could consider splitting the
priority information out into the various drivers. For now, though, they
have no way of knowing that, so it's easier to put them in an easy to find
table.
Also, move the checkversion() call in vmstat(8) to a more logical place.
Thanks to Bruce and David O'Brien for suggestions, for reviewing this, and
for putting up with the long time it has taken me to commit it. Bruce did
object somewhat to the central priority table (he would rather the
priorities be distributed in each driver), so his objection is duly noted
here.
Reviewed by: bde, obrien
to an architecture-specific value defined in <machine/elf.h>. This
solves problems on large-memory systems that have a high value for
MAXDSIZ.
The load address is controlled by a new macro ELF_RTLD_ADDR(vmspace).
On the i386 it is hard-wired to 0x08000000, which is the standard
SVR4 location for the dynamic linker.
On the Alpha, the dynamic linker is loaded MAXDSIZ bytes beyond
the start of the program's data segment. This is the same place
a userland mmap(0, ...) call would put it, so it ends up just below
all the shared libraries. The rationale behind the calculation is
that it allows room for the data segment to grow to its maximum
possible size.
These changes have been tested on the i386 for several months
without problems. They have been tested on the Alpha as well,
though not for nearly as long. I would like to merge the changes
into 3.1 within a week if no problems have surfaced as a result of
them.
attempt to optimize forks but were essentially given-up on due to
problems and replaced with an explicit dup of the vm_map_entry structure.
Prior to the removal, they were entirely unused.
