the past, since it returns to the old system of allocating mbufs out of
a private area rather than using the kernel malloc(). While this may seem
like a backwards step to some, the new allocator is some 20% faster than
the old one and has much better caching properties.
Written by: John Wroclawski <jtw@lcs.mit.edu>
very busy servers (eg: news, web). This is an interaction between
embryonic processes that have not yet finished forking, and happen to
cause the kernel VM space to grow, hitting the uninitialised variable.
It was possible for this to strike at any time, depending on the size of
your kernel and load patterns. One machine had paniced occasionally
when cron launches a job since before the 2.1 release.
If you had "options DIAGNOSTIC", you may have seen references to bogus
addresses like 0xdeadc142 and the like.
This is a minimal change to fix the problem, it will probably be done
better by reordering p_vmspace to be in the startzero section, but it
becomes harder to validate then.
It's been vulnerable since pmap.c rev 1.40 (Jan 9, 1995), so it's been a
cause of problems since well before 2.0.5. This was when the merged
VM/buffer cache and the dynamic growing kernel VM space were first
committed. This probably fixes a few of PR's.
and B_READ before writing. This was was fatal. They also broke the
clearing of B_INVAL before doing i/o. This didn't actually matter.
Submitted by: mostly by joerg
compatibility slice. They were forgotten on last-close and then
creating them on first-open failed.
Devfs entries for slices other than the one containing the root file
system are still invisible unless you open a non-devfs inode on the
slice.
forked child to be dissociated from the parent).
Cleanup fork1(), implement vfork() and fork() in terms of rfork() flags.
Remove RFENVG, RFNOTEG, RFCNAMEG, RFCENVG which are Plan9 specific and cannot
possibly be implemented in FreeBSD.
Renumbered the flags to make up for the removal of the above flags.
Reviewed by: peter, smpatel
Submitted by: Mike Grupenhoff <kashmir@umiacs.umd.edu>
was duplicated until the canq filled up, and write() normally returned 0.
This case is apparently rare. It was reported for Jove's shell buffer in
PR 1130.
Make the SA_NODEFER handling more correct, previously if you called
sigaction to set a handler and had SA_NODEFER set, and manually masked
the signal itself in sa_mask, and when you read the settings back later,
you'd find SA_NODEFER incorrectly cleared.
Pointed out by: bde
the process's memory, it was possible for the procfs_domem() call to
return a residual leftover, but with no errno. Since this is no good for
ptrace which ignored the the residual, remap a leftover amount into an
errno rather than fooling the caller into thinking it was successful when
in fact it was not.
Submitted by: bde (a very long time ago :-)
Made the devfs `fd' devices bug for bug compatible with the ones created
by MAKEDEV:
- ownership is bin.bin, not root.wheel, except for std*. The devfsext
interface doesn't seem to allow specifying the ownership of /devfs/fd,
so it's still incompatible.
- std* aren't links to fd/[0-2].
Added scsi control devices.
Converted almost everything that I changed to use devfs_add_devswf()
and verbose id macros.
st.c:
Renamed enrst* to erst* since that's what the current name is (enrst
seems to be an old name).
vm_offset_t is currently unsigned long but should probably be plain
unsigned for i386's to match the choice of minimal types to represent
for fixed-width types in Lite2. Anyway, it shouldn't be assumed
to be unsigned long.
I only fixed the type mismatches that were detected when I changed
vm_offset_t to unsigned. Only pointer type mismatches were detected.
but didn't actually do anything with it (*blush*).
This should fix bde's test case where the test program set SA_RESETHAND
and when reading it back, it was gone.
Tweak/optimize SA_NODEFER so that the implementation is a little simpler
and does not incur (slight) overhead for every signal at delivery time.
files are off the vendor branch, so this should not change anything.
A "U" marker generally means that the file was not changed in between
the 4.4Lite and Lite-2 releases, and does not need a merge. "C" generally
means that there was a change.
[note new unused (in this form) syscalls.conf, to be 'cvs rm'ed]
the obsolete soqinsque and soqremque functions as well as collapsing
so_q0len and so_qlen into a single queue length of unaccepted connections.
Now the queue of unaccepted & complete connections is checked directly
for queued sockets. The new code should be functionally equivilent to
the old while being substantially faster - especially in cases where
large numbers of connections are often queued for accept (e.g. http).
Compile and link a new kernel, that will give native ELF support, and
provide the hooks for other ELF interpreters as well.
To make native ELF binaries use John Polstras elf-kit-1.0.1..
For the time being also use his ld-elf.so.1 and put it in
/usr/libexec.
The Linux emulator has been enhanced to also run ELF binaries, it
is however in its very first incarnation.
Just get some Linux ELF libs (Slackware-3.0) and put them in the
prober place (/compat/linux/...).
I've ben able to run all the Slackware-3.0 binaries I've tried
so far.
(No it won't run quake yet :)
residing in a buffer that had been dirtied by a process was being
handled incorrectly. The pages were mistakenly placed into the
cache queue. This would likely have the effect of mmaped page modifications
being lost when I/O system calls were being used simultaneously to
the same locations in a file.
Submitted by: davidg
on in the FreeBSD development, I had made a global lock around the
rlist code. This was bogus, and now the lock is maintained on a
per resource list basis. This now allows the rlist code to be used for
almost any non-interrupt level application.
linux binaries from the *BSD a.out loader. This is a hack, but lets me run
static NetBSD binaries. Dynamic binaries are a much bigger problem because
the shared libraries would conflict with our native libraries, so a
/compat/netbsd alternate namespace and translation would be needed.
netscape-2.0 for Linux running all the Java stuff. The scrollbars are now
working, at least on my machine. (whew! :-)
I'm uncomfortable with the size of this commit, but it's too
inter-dependant to easily seperate out.
The main changes:
COMPAT_LINUX is *GONE*. Most of the code has been moved out of the i386
machine dependent section into the linux emulator itself. The int 0x80
syscall code was almost identical to the lcall 7,0 code and a minor tweak
allows them to both be used with the same C code. All kernels can now
just modload the lkm and it'll DTRT without having to rebuild the kernel
first. Like IBCS2, you can statically compile it in with "options LINUX".
A pile of new syscalls implemented, including getdents(), llseek(),
readv(), writev(), msync(), personality(). The Linux-ELF libraries want
to use some of these.
linux_select() now obeys Linux semantics, ie: returns the time remaining
of the timeout value rather than leaving it the original value.
Quite a few bugs removed, including incorrect arguments being used in
syscalls.. eg: mixups between passing the sigset as an int, vs passing
it as a pointer and doing a copyin(), missing return values, unhandled
cases, SIOC* ioctls, etc.
The build for the code has changed. i386/conf/files now knows how
to build linux_genassym and generate linux_assym.h on the fly.
Supporting changes elsewhere in the kernel:
The user-mode signal trampoline has moved from the U area to immediately
below the top of the stack (below PS_STRINGS). This allows the different
binary emulations to have their own signal trampoline code (which gets rid
of the hardwired syscall 103 (sigreturn on BSD, syslog on Linux)) and so
that the emulator can provide the exact "struct sigcontext *" argument to
the program's signal handlers.
The sigstack's "ss_flags" now uses SS_DISABLE and SS_ONSTACK flags, which
have the same values as the re-used SA_DISABLE and SA_ONSTACK which are
intended for sigaction only. This enables the support of a SA_RESETHAND
flag to sigaction to implement the gross SYSV and Linux SA_ONESHOT signal
semantics where the signal handler is reset when it's triggered.
makesyscalls.sh no longer appends the struct sysentvec on the end of the
generated init_sysent.c code. It's a lot saner to have it in a seperate
file rather than trying to update the structure inside the awk script. :-)
At exec time, the dozen bytes or so of signal trampoline code are copied
to the top of the user's stack, rather than obtaining the trampoline code
the old way by getting a clone of the parent's user area. This allows
Linux and native binaries to freely exec each other without getting
trampolines mixed up.
queue type is not set to QUEUE_NONE. This appears to have
caused a hang bug that has been lurking.
2) Fix bugs that brelse'ing locked buffers do not "free" them, but the
code assumes so. This can cause hangs when LFS is used.
3) Use malloced memory for directories when applicable. The amount
of malloced memory is seriously limited, but should decrease the
amount of memory used by an average directory to 1/4 - 1/2 previous.
This capability is fully tunable. (Note that there is no config
parameter, and might never be.)
4) Bias slightly the buffer cache usage towards non-VMIO buffers. Since
the data in VMIO buffers is not lost when the buffer is reclaimed, this
will help performance. This is adjustable also.
is <sys/unistd.h>, with the prototype in <unistd.h>. sys/unistd.h
is visible to the kernel compile, and is #included by unistd.h.
Also, I missed a reference to a static int in the midst of my other diffs.
kern_fork.c: add the tiny bit of code for rfork operation.
kern/sysv_*: shmfork() takes one less arg, it was never used.
sys/shm.h: drop "isvfork" arg from shmfork() prototype
sys/param.h: declare rfork args.. (this is where OpenBSD put it..)
sys/filedesc.h: protos for fdshare/fdcopy.
vm/vm_mmap.c: add minherit code, add rounding to mmap() type args where
it makes sense.
vm/*: drop unused isvfork arg.
Note: this rfork() implementation copies the address space mappings,
it does not connect the mappings together. ie: once the two processes
have split, the pages may be shared, but the address space is not. If one
does a mmap() etc, it does not appear in the other. This makes it not
useful for pthreads, but it is useful in it's own right for having
light-weight threads in a static shared address space.
Obtained from: Original by Ron Minnich, extended by OpenBSD
fixes for previous version of new pipes from Bruce Evans. This
new version:
Supports more properly the semantics of select (BDE).
Supports "OLD_PIPE" correctly (kern_descrip.c, BDE).
Eliminates incorrect EPIPE returns (bash 'pipe broken' messages.)
Much faster yet, currently tuned relatively conservatively -- but now
gives approx 50% more perf than the new pipes code did originally.
(That was about 50% more perf than the original BSD pipe code.)
Known bugs outstanding:
No support for async io (SIGIO). Will be included soon.
Next to do:
Merge support for FIFOs.
Submitted by: bde
The code outputs the dc then calls the device specific externalize
routines to fill in the dc_data area. The old code assumed that dc_data
started one byte from the end of the dc, but with the compiler optimizing
alignment and padding, this isn't always the case. Do an explicit
&(dc.dc_data) - &dc. This fixes lsdev -c which must have been broken
for some time.
1) The calculation didn't account for NMBCLUSTERS, so if a large number of
clusters was specified, it would leave little or no space for kernel
malloc.
2) It was bogusly restricted to v_page_count. This doesn't take into
account the sparseness of the malloc area and would have caused
problems on machines with small amounts of memory. It should probably
instead be changed to set the malloc limit to be constrained by
the amount of memory, but I didn't do this.
First attempt at creating devfs entries for sliced devices. Doesn't
quite work yet, so the heart of it is disabled.
Added bdev and cdev args to dsopen().
Create devfs entries in dsopen() and (unsuccessfully) attempt to make
them go away at the right times. DEVFS is #undefed at the start so
that this shouldn't cause problems.
PT_ATTACH/PT_DETACH implemented now and fully operational.
PT_{GET|SET}{REGS|FPREFS} implemented now, using code shared with procfs
PT_{READ|WRITE}_{I|D} now uses code shared with procfs
ptrace opcodes now fully permission checked, including ownerships.
doing an operation to the u-area on a swapped process should no longer
panic.
running gdb as root works for me now, where it didn't before.
general cleanup..
Note, that this has some tightening of permissions/access checks etc.
Some of these may be going too far.. In particular, the "owner" of the
traced process is enforced. The process that created or attached to
the traced process is now the only one that can "do" things to it.
Speed up for vfs_bio -- addition of a routine bqrelse to greatly diminish
overhead for merged cache.
Efficiency improvement for vfs_cluster. It used to do alot of redundant
calls to cluster_rbuild.
Correct the ordering for vrele of .text and release of credentials.
Use the selective tlb update for 486/586/P6.
Numerous fixes to the size of objects allocated for files. Additionally,
fixes in the various pagers.
Fixes for proper positioning of vnode_pager_setsize in msdosfs and ext2fs.
Fixes in the swap pager for exhausted resources. The pageout code
will not as readily thrash.
Change the page queue flags (PG_ACTIVE, PG_INACTIVE, PG_FREE, PG_CACHE) into
page queue indices (PQ_ACTIVE, PQ_INACTIVE, PQ_FREE, PQ_CACHE),
thereby improving efficiency of several routines.
Eliminate even more unnecessary vm_page_protect operations.
Significantly speed up process forks.
Make vm_object_page_clean more efficient, thereby eliminating the pause
that happens every 30seconds.
Make sequential clustered writes B_ASYNC instead of B_DELWRI even in the
case of filesystems mounted async.
Fix a panic with busy pages when write clustering is done for non-VMIO
buffers.
Add more features to the one remaining to handle the job:
+ signed quantity.
# alternate format
- left padding
* read width as next arg.
n numeric in (argument specified) default radix.
Fix the DDB debugger to use these.
Use vprintf in debug routine in pcvt.
The warnings from gcc may become more wrong and intolerable because
of this.
Warning: I have not checked the entire source for unsupported or
changed constructs, but generally belive that there are only a few.
Suggested by: bde
sysv_ipc.c: add stub functions that either simply return (for the hooks
in kern_fork/kern_exit) or log() a messgae and call enosys() (for the
syscalls). sysv_ipc.c will become "standard" in conf/files and has
#ifs for all the permutations.
are about to go in. This is to fix the problem with the ibcs2 and linux
lkm's not being able to call the sysv ipc functions unless the build is
modified.
the range [210:260] by sweeping the problem under the rug. This change
has the following effects:
1) A new MIB variable in the kern branch is defined to allow modification
of the socket buffer layer's ``wastage factor'' (which determines how
much unused-but-allocated space in mbufs and mbuf clusters is allowed
in a socket buffer).
2) The default value of the wastage factor is changed from 2 to 8.
The original value was chosen when MINCLSIZE was 7*MLEN (!), and is not
appropriate for an environment where MINCLSIZE is much less.
The real solution to this problem is to scrap both mbufs and sockbufs
and completely redesign the buffering mechanism used at both levels.
sysctl handler (ouch!)
Add a "const" qualifier to the source of the copyin() and copyout()
functions - the other const warning in kern_sysctl.c was silenced when
copyout was declared as having a const source.. (which it is)
just like on SVR4.
This has no effect on any current programs in our source, but makes
the use of SVR4 code a little easier. There is no code or implementation
cost in the kernel.. This two-line change merely sets the modes on the ends
of the pipes to be bidirectional. There are no other changes.
looking at a high resolution clock for each of the following events:
function call, function return, interrupt entry, interrupt exit,
and interesting branches. The differences between the times of
these events are added at appropriate places in a ordinary histogram
(as if very fast statistical profiling sampled the pc at those
places) so that ordinary gprof can be used to analyze the times.
gmon.h:
Histogram counters need to be 4 bytes for microsecond resolutions.
They will need to be larger for the 586 clock.
The comments were vax-centric and wrong even on vaxes. Does anyone
disagree?
gprof4.c:
The standard gprof should support counters of all integral sizes
and the size of the counter should be in the gmon header. This
hack will do until then. (Use gprof4 -u to examine the results
of non-statistical profiling.)
config/*:
Non-statistical profiling is configured with `config -pp'.
`config -p' still gives ordinary profiling.
kgmon/*:
Non-statistical profiling is enabled with `kgmon -B'. `kgmon -b'
still enables ordinary profiling (and distables non-statistical
profiling) if non-statistical profiling is configured.
int shmget(key_t key, int size, int shmflg);
If the 'key' has already existed in the system and set 'shmflg'
as '(IPC_CREAT|IPC_EXC)', then shmget() must return the error 'EEXIST'.
Submitted by: m_tanaka@pa.yokogawa.co.jp (Mihoko Tanaka)
See the comments for addupc_intr() and the NetBSD implementation.
We use dummy versions of fuswintr() and susiwintr(), so addupc_intr()
always pushes the work to trap() (this is inefficient), and trap()
calls the special i386 function addupc() instead of addupc_task().
addupc() is more efficient than addupc_intr(), so some of the lost
efficiency is recovered. However, addupc() may be broken on plain
i386's since it doesn't check for write permission like copyout().
redistribute a few last routines to beter places and shoot the file
I haven't act actually 'deleted' the file yet togive people time
to
have done a config.. I.e. they are likely to have done one in a week or so
so I'll remove it then..
it's now empty.
makes the question of a USL copyright rather moot.
overflows.
It sure would be nice if there was an unmapped page between the PCB and
the stack (and that the size of the stack was configurable!). With the
way things are now, the PCB will get clobbered before the double fault
handler gets control, making somewhat of a mess of things. Despite this,
it is still fairly easy to poke around in the overflowed stack to figure
out the cause.
Unstaticize a function in scsi/scsi_base that was used, with an undocumented
option.
My last count on the LINT kernel shows:
Total symbols: 3647
unref symbols: 463
undef symbols: 4
1 ref symbols: 1751
2 ref symbols: 485
Approaching the pain threshold now.
*' instead of caddr_t and it isn't optional (it never was). Most of the
netipx (and netns) pr_ctlinput functions abuse the second arg instead of
using the third arg but fixing this is beyond the scope of this round
of changes.
(TTMAXHIWAT + OBUFSIZ + 100) in case someone changes OBUFSIZ. 200
was to allow 100 above high water for ordinary writes and another
100 for kernel printfs.
Increased the reserved output queue count from 512 to the maximum
output queue count. This prevents exhaustion of clists and increases
the output throughput for 8 cy lines by almost a factor of 2 (on
a system where there aren't many other open ttys so clists become
exhausted after about 4 active lines (or earlier if TTMAXHIWAT is
increased :-]).
ttwrite() behaves very badly when clists are exhausted:
(1) it sleeps on lbolt instead of on TSA_OLOWAT(tp).
This could be fixed adequately by sleeping on TSA_OLOWAT(tp).
The nonzero reserved count guaratees that space will become
available independent of other ttys, and a reserved count
of 512 is barely enough for efficiency.
(2) it drops output if space runs out in the middle of special
output processing. This is too hard to fix without hardening
the reserved count. The watermark processing guarantees that
space doesn't run out only if the advertised space is guaranteed.
Increasing the reserved output queue count defeats the point of
dynamic allocation of clists. Previously, about 2K of memory per
tty was reserved (the raw queue was already reserved). Now, about
3.5K is reserved. Reserving everything would take a whole 0.5K
more.
ttyflush() might be called from an interrupt handler. This fixes
panics in IXOFF mode at the cost of more failures to send the START
character to exit from IXOFF mode.
LINT: add a couple of new/missing/undocumented options
files.i386: add linux code so that you can compile a kernel with static
linux emulation ("options LINUX")
i386/*: use #if defined(COMPAT_LINUX) || defined(LINUX) to enable static
support of linux emulation (just like "IBCS2" makes ibcs2 static)
The main thing this is going to make obvious, is that the LINUX code
(when compiled from LINT) has a lot of warnings, some of which dont look
too pleasant..
seems to work hre just fine though I can't check every file
that changed due to limmited h/w, however I've checked enught to be petty
happy withe hte code..
WARNING... struct lkm[mumble] has changed
so it might be an idea to recompile any lkm related programs
metadata and VBLK type devices. The code is currently mostly disabled,
and a work-around has been added to disabled attempted clustered writes
for VBLK type device buffers. Clustered write of meta-data is currently
a work in progress.
"getblk" hang. The B_WANTED flag was being cleared gratuitously,
also the optimization of gbincore for ignoring the B_INVAL flag was
incorrect. There is no place in the code where buffers are on the
hash list that are B_INVAL and not B_BUSY.
Removed old aliases d_rdwr_t and d_ttycv_t for d_read_t/d_write_t and
d_devtotty_t.
Sorted declarations of switch functions into switch order.
Removed duplicated comments and declarations of nonexistent switch
functions.
This is now in line with NetBSD as well..
Note that once this series of commits is finished, you must recompile
libkvm, then ps and maybe 'w'. If you are running the recently imported
sendmail-8.7, you should recompile that too (src/conf.c at least).
most devsw referenced functions are now static, as they are
in the same file as their devsw structure. I've also added DEVFS
support for nearly every device in the system, however
many of the devices have 'incorrect' names under DEVFS
because I couldn't quickly work out the correct naming conventions.
(but devfs won't be coming on line for a month or so anyhow so that doesn't
matter)
If you "OWN" a device which would normally have an entry in /dev
then search for the devfs_add_devsw() entries and munge to make them right..
check out similar devices to see what I might have done in them in you
can't see what's going on..
for a laugh compare conf.c conf.h defore and after... :)
I have not doen DEVFS entries for any DISKSLICE devices yet as that will be
a much more complicated job.. (pass 5 :)
pass 4 will be to make the devsw tables of type (cdevsw * )
rather than (cdevsw)
seems to work here..
complaints to the usual places.. :)
single typedef) is now declared in <sys/types.h>.
This is the first of 4 commits that will remove some excessive
includes of vm*.h and user.h. The total speed improvement isn't
as large as I first thought. `make depend; make' for LINT only
improved from 2180 seconds to 2108 seconds user time.
Move a lot of variables home to their own code (In good time before xmas :-)
Introduce the string descrition of format.
Add a couple more functions to poke into these marvels, while I try to
decide what the correct interface should look like.
Next is adding vars on the fly, and sysctl looking at them too.
Removed a tine bit of defunct and #ifdefed notused code in swapgeneric.
Obtained from: NetBSD as well (He submitted it there too)
make sure that teh shm region is beyond the sum of the text and data segs
as it was big progs could collide with the shm region.
but also IT ACTUALLY WORKS!
FreeBSD with options JREMOD now runs with no entries in the devsw tables
prior to the devices puting their own entries there..
Thanks to bde and terry for thoughts and comments.
next stop 'Real' devfs support in devices.
add a few safety checks in specfs because
now it's possible to get entries in [cd]devsw[] which are ALL NULL
so it's better to discover this BEFORE jumping into the d_open() entry..
more check to come later.. this getsthe code to the stage where I
can start testing it, even if I haven't caught every little error case...
I guess I'll find them quick enough..
That's EVERY SINGLE driver that has an entry in conf.c..
my next trick will be to define cdevsw[] and bdevsw[]
as empty arrays and remove all those DAMNED defines as well..
Each of these drivers has a SYSINIT linker set entry
that comes in very early.. and asks teh driver to add it's own
entry to the two devsw[] tables.
some slight reworking of the commits from yesterday (added the SYSINIT
stuff and some usually wrong but token DEVFS entries to all these
devices.
BTW does anyone know where the 'ata' entries in conf.c actually reside?
seems we don't actually have a 'ataopen() etc...
If you want to add a new device in conf.c
please make sure I know
so I can keep it up to date too..
as before, this is all dependent on #if defined(JREMOD)
(and #ifdef DEVFS in parts)
self-decompressing ram disk that I'm fiddling with..
(Note, this depends on the various syscalls having correctly set uio_segflag
before calling physio - I've checked and they look correct.)
Convert the remaining sysctl stuff to the new way of doing things.
the devconf stuff is the reason for the large number of files.
Cleaned up some compiler warnings while I were there.
1) Make cluster buffer list be a non-malloced chain. This eliminates
yet another 'evil' M_WAITOK and generally cleans up the code.
2) Fix write clustering for ext2fs. It was just broken. Also, ffs
clustering had an efficiency problem that more bawrites were happening
than should have been.
3) Make changes to buf.h to support the above, plus remove b_pfcent
at the request of David Greenman.
Reviewed by: davidg (partially)
- don't allow invalid timevals.
- normalize timevals as they are built - don't call timevaladd() with
a possibly invalid timeval and normalize the result.
Fixed a warning.
