Somebody should make a mib variable for it.
Just now it is pointless to dump the kernel, since we have nothing which
can read the dump.
Furthermore is should never be the default to dump.
options DODUMP
will enable dumps.
for all reasonable HZ's. HZ > 1000 doesn't work because of sloppy
conversions in hzto() (division by (tick / 1000) == 0). This was
fixed in 1.1.5.
Eliminate some extern declarations by including the appropriate header
files that now contain appropriate declarations.
doesn't have to calculate it every call.
Rename `timer0_prescale' to `timer0_prescaler_count' and maintain it
correctly. Previously we lost a few 8253 cycles for every "prescaled"
clock interrupt, and the lossage grows rapidly at 16 KHz. Now we
only lose a few cycles for every standard clock interrupt.
Rename `*_divisor' to `*_max_count'.
Do the calculation of TIMER_DIV(rate) only once instead of 3 times each
time the rate is changed.
Don't allow preposterously large interrupt rates. Bug fixes elsewhere
should allow the system to survive rates that saturate the system, however.
Clean up declarations.
Include <machine/clock.h> to check our own declarations.
for it is incomplete and buggy. There is no problem unless Xintr0()
is reentered or should be reentered, but high clock interrupt
frequencies for pcaudio cause Xintr0() to be reentered (or clock
ticks to be lost when Xintr0() should have been reentered but
wasn't), and we lose little by delaying the call to softclock().
Move declarations related to the clock driver to clock.h.
Move declarations related to the npx driver to npx.h.
Clean up the remaining declarations.
I know that many of these entries are bogus and need to be revisited,
but let's get the tree working again for now and then do a pass through
looking at all the __FreeBSD__ entries, shall we?
Cosmetic.
Return from trap() if trap_fatal() returns. trap_fatal() isn't
fatal if you have ddb. Returning from trap() is usually the right
thing to do and much better than falling through.
Build a dummy frame at the top of tmpstk to help debuggers trace the stack
when the system is idle.
swtch.s: idle():
Initialize the frame pointer so that debuggers don't try to trace a bogus
stack.
Load the frame pointer, load the stack pointer and switch out the old
stack in the unique order that never leaves one of the pointers pointers
invalid so that debuggers can trace idle(). Disabling interrupts
provides sufficient validity for normal operation, but debuggers use
(trace) traps.
Changed the fifth parameter to register_intr() from u_int mask into
u_int *maskptr in preparation for new features (shared interrupts and
removable devices, eg. for PCMCIA).
Changed the fifth parameter to register_intr() from u_int mask into
u_int *maskptr in preparation for new features (shared interrupts and
removable devices, eg. for PCMCIA).
of memory to work without running out of kernel VM (and increasing it to
even more than it is now (96MB) is out of the question. Changed bufpages
calculation to allocation a little less bufer cache (16% of mem-2MB instead
of 20%); this is simply a better figure for most systems.
text. Fixed rounding bug that caused the last page of kernel text to be
read/write instead of read-only. This is important now that tmpstk can
crash into it. Removed +4 bias of tmpstk because it screws up ddb's
ability to traceback correctly.
and all SCSI devices (except that it's not done quite the way I want). New
information added includes:
- A text description of the device
- A ``state''---unknown, unconfigured, idle, or busy
- A generic parent device (with support in the m.i. code)
- An interrupt mask type field (which will hopefully go away) so that
. ``doconfig'' can be written
This requires a new version of the `lsdev' program as well (next commit).
explanation. More doc needed, but not hard to do, if you want to.
A big hand to Martin Renters for the netboot program !
Anybody want to compete on who can "make world" in the shortest
amount of time ? I have 127 i486DX2/66 and 5 P60's I can use
now. And 3 times 66 Gb file servers to support it... :->
Anyway, NFS will be standard in the GENERIC kernel now, so that
people can use the bin-tarball to set up shop.
drivers have a chance to change their IRQ before it is checked.
This was implemented in revision 1.21 and broken in revision 1.26.
Drivers that can change their IRQ should probably be configured
with "irq ?".
else has been probed. This feature could go away again, if we can curb the
problem another way.
if_ed.c, syscons.c: Set the above flag. ed# because it needs it, syscons
because it looks stupid to "detect" the display you have already filled up
with text :-)
bt742a.c: Check bt_cmd() return-val during probe, thus failing on adaptec's.
Also silenced various printf's during the probe.
isa.c: Probe devices with the above flag set before the rest. Reduce the
number of "conflict" messages per device to one.
***
Please test the GENERIC-kernel now, if nobody can make it fail, GENERICAH
and GENERICBT has a finite and short life-expectancy...
***
scheme of things, so I've changed them to be more appropriate. page in/ous
are now associated with the pager that did them. Nuked v_fault as the
only fault of interest that wouldn't be already counted in v_trap is a VM
fault, and this is counted seperately.
2) Implemented most of the remaining counters and corrected the counting of
some that were done wrong. They are all almost correct now...just a few
minor ones left to fix.
"APM" macro.
machdep.c: Made the APM-descriptors unconditional.
Bruce: if these still conflict with your debugger, please put in a reservation
for your debugger. These three desc. can be anywhere, as long as they are
contiguous, so just move them as needed.
pmap.c: tons of unused vars zapped, various other warnings silenced.
trap.c: unused vars zapped.
vm_machdep.c: A wrong argument, which by chance did the right thing, was
corrected.
that and when it does it will be done differently.
2. The kernel now does a frame setup on entry so it ``looks'' like a
real function call. This will be needed by future boot code and
debuggers.
3. Clean up stack offsets to all be in decimal and use %ebp when copying
parameters in from the boot code.
4. Implement version 1 of the uniform boot code passing mechanism with
support for kernelname passing and nfs_diskless structure passing.
5. Document the 3 different ways the kernel is called depending on what code
is calling it.
must #define NFS before including <sys/mount.h> to pick up some of
the definitions needed for struct diskless. Be sure to undef it after this
so you do not effect other code.
This is kinda sick, but it does the job. Problem found by davidg.
2. New detection code so we know what boot code called us.
3. Remove old DISKLESS support code and halt if we are called by that boot
code as it will NOT work with the new nfs_diskless structure.
This is really in preperation for new boot code and new diskless support.
Reviewed by: davidg
including files. vector.s sometimes left the data section misaligned
(depending on the configuration) so all the time-critical globals in icu.s
were sometimes misaligned.
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.
matter, but similar bogusness in npx.c causes compiling without -O to fail.
Use __volatile in all asms.
Parenthesize macro args.
Change the names of the macros to avoid namespace pollution.
Remove unnecessary "#ifdef __i386__".
Sort #defines.
Add comments.
have got the following:
Back out the changes in the previous revision. Function-like macros
were replaced by compound statements that work in less contexts.
Unoformize idempotency #ifdef.
Restore the simple leap year calculation as a macro and document it so
that it doesn't become complicated again. The simple version works
for all leap years covered by 32-bit time_t's. The complicated version
doesn't work for all leap years covered by 64-bit time_t's since among
other reasons, the solar system is not stable for long enough.
Fix declarations.
Nuke spinwait().
This code is mostly taken from the 1.1 port (which was in turn taken from
Dave Mills's kern.tar.Z example). A few significant differences:
1) ntp_gettime() is now a MIB variable rather than a system call. A few
fiddles are done in libc to make it behave the same.
2) mono_time does not participate in the PLL adjustments.
3) A new interface has been defined (in <machine/clock.h>) for doing
possibly machine-dependent things around the time of the clock update.
This is used in Pentium kernels to disable interrupts, set `time', and
reset the CPU cycle counter as quickly as possible to avoid jitter in
microtime(). Measurements show an apparent resolution of a bit more than
8.14usec, which is reasonable given system-call overhead.
Removed inb function since it's more correctly in pio.h
Copied write_eflags and read_eflags over from npx.c
(Some changes to the macros suggested by Bruce were not made at this
time since his suggestions probably apply to all the macros and
these inlined/macro definitions need a lot of cleaning up at some
point in the future.)
Reviewed by: Bruce
Fix from Bruce Evans. There were missing sets of parantheses:
1. The checks for the standard data selectors were botched, so %ss == 0
and probably %cs == 0 were allowed. A fix is enclosed. The checks
for the standard selectors could be omitted without losing anything
since the standard selectors pass the valid_ldt_sel() tests.
320x200 256col VGA. This is nessesary for the iBCS stuff to work right.
(And we get the benefit of more video modes). Uses the videocard BIOS
to optain mode tables.
Added a "green" saver, switches off the syncs for "green" monitors.
Reviewed by:
Submitted by:
Obtained from:
don't hard-code netisr values in icu.s, but rather, use an array of
function pointers and set them all up in machdep.c for statically-linked
protocol families. (This will eventually be done differently.)
otherwise the machine will overflow the stack in a recursive fault loop
(causing the machine to spontaneously reboot because of the stack fault
that ultimately happens).
Submitted by: Inspired by Bruce Evans, but this change is different
than what he suggested.
if KERNEL is not defined. lib/msun/i387/*.S include asmacros.h to
get the definitions of ENTRY(), etc. This is bogus since asmacros.h
is only supposed to give definitions suitable for the kernel. The
current definitions for the kernel almost worked but are missing
the ".type" declarations. This caused the linker to print warnings
about doubtful relocations for almost anything linked to libm[sun].
Uniformize name and use of idempotence identifier.
the Mach/i386 version of the BSD/vax(?) <machine/psl.h>. The Mach
version has slightly better names for many macros but is now out of
date and little used. It was originally used even less (for spelling
PSL_T as EFL_TF in <machine/db_machdep.h>).
negation whenever we access memory between 640k and 1M.
Original code from NetBSD 1.0-BETA. The exact origins are unclear but
Theo de Raadt, Charles, and Michael V. may have contributed to it.
Submitted by: pst
Submitted by:
1) if_ie.c:
Changed a printf and put a space in it. Formerly the "<3C507>"
confused the syslog. He tried to see that as the priority to
log that message.
2) isa_device.h:
Changed the iobase variable from short to u_short. EISA
Adresses can go up to 0xf000 and the sign extension doesn't
look good in the probe output. Example:
ep1 at 0xffff8000-0xffff8000f is not good :-), i like more a
ep1 at 0x8000-0x8000f.
3) isa.c:
Changed a string constant from "probe" to "prob", it gets
later already an "ed" tagged on the end.
with BOUNCE_BUFFERS. This is more intuitive, and is better for future
multiplatform support. Added BOUNCE_BUFFERS option to the GENERIC and
LINT kernel config files.
in your kernel config now).
2) Added ps ddb function from 1.1.5. Cleaned it up a bit and moved into its
own file.
3) Added \r handing in db_printf.
4) Added missing memory usage stats to statclock().
5) Added dummy function to pseudo_set so it will be emitted if there
are no other pseudo declarations.
Changed u_int_inb to just inb and deleted define.
The code generated is identical to that generated with the cast so
the problem was obviously fixed at some point after gcc 1.4
Reviewed by:
Submitted by:
hosing syscons. Doesn anyone know anything about this
or can we just delete it now?
/*
* This roundabout method of returning a u_char helps stop gcc-1.40 from
* generating unnecessary movzbl's.
*/
#ifdef disable_for_gcc-2_6_0
#define inb(port) ((u_char) u_int_inb(port))
#endif
static inline u_int
u_int_inb(u_int port)
{
u_char data;
/*
* We use %%dx and not %1 here because i/o is done at %dx and
not at
* %edx, while gcc-2.2.2 generates inferior code (movw instead
of movl)
* if we tell it to load (u_short) port.
*/
__asm __volatile("inb %%dx,%0" : "=a" (data) : "d" (port));
return data;
}
Reviewed by:
Submitted by:
machdep.c:
Changed printf's a little and call vfs_unmountall() if the sync was
successful.
cd9660_vfsops.c, ffs_vfsops.c, nfs_vfsops.c, lfs_vfsops.c:
Allow dismount of root FS. It is now disallowed at a higher level.
vfs_conf.c:
Removed unused rootfs global.
vfs_subr.c:
Added new routines vfs_unmountall and vfs_unmountroot. Filesystems
are now dismounted if the machine is properly rebooted.
ffs_vfsops.c:
Toggle clean bit at the appropriate places. Print warning if an
unclean FS is mounted.
ffs_vfsops.c, lfs_vfsops.c:
Fix bug in selecting proper flags for VOP_CLOSE().
vfs_syscalls.c:
Disallow dismounting root FS via umount syscall.
2. Hack.
Hack is to define RCSID() to null macro so that new msun stuff
will compile. This does NOT belong here, and I DON'T want it to
stay, I just need to put this here for now to enable msun and we need
to talk about what our RCSID story is supposed to be. We talked about
supporting RCSID() one day, and everyone seemed to like the idea
reasonably well of making it a macro you could just no-op this way,
but we never did anything. Now I see that JTCs code has it and I'm
loath to remove it or do anything until we've discussed it some more.
Well, so how about it? What's our story vis-a-vis RCSID() going to
be?
Submitted by: jkh
- Delete redundant declarations.
- Add -Wredundant-declarations to Makefile.i386 so they don't come back.
- Delete sloppy COMMON-style declarations of uninitialized data in
header files.
- Add a few prototypes.
- Clean up warnings resulting from the above.
NB: ioconf.c will still generate a redundant-declaration warning, which
is unavoidable unless somebody volunteers to make `config' smarter.
/usr/src/sys/i386/isa/clock.c:
o Garrett's statclock changes.
o Wire xxxintr, not Vclk.
o Wire using register_intr(), not setidt().
/usr/src/sys/i386/isa/icu.s:
o Garrett's statclock changes.
o Removed unused variable high_imask.
o Fake int 8 for rtc as well as int 0 for clk. Required for kernel
profiling with statclock, harmless otherwise.
/usr/src/sys/i386/isa/isa.c:
o Allow isdp->id_irq and other things in *isdp to be changed by
probes. Changing interrupts later requires direct calls to
register_intr() and unregister_intr() and more care.
ALLOW_CONFLICT_* is brought over from 1.1.5, except
ALLOW_CONFLICT_IRQ is not supported. IRQ conflict checking is
delayed until after probing so that drivers can change the IRQ
to a free one; real conflicts require more cooperation between
drivers to handle.
o Too many details to list.
o This file requires splitting and a lot more work.
/usr/src/sys/i386/isa/isa_device.h:
o Declare more things more completely.
/usr/src/sys/i386/isa/sio.c:
o Prepare to register interrupt handlers as fast.
/usr/src/sys/i386/isa/vector.s:
o Generate entry code for 16 fast interrupt handlers and 16 normal
interrupt handlers. Changed some constants to variables:
# $unit is now intr_unit[intr]. Type is int. Someday it should
be a cookie suitable for the handler (e.g., a struct com_s for
sio).
# $handler is now intr_handler[intr].
# intrcnt_actv[id_num] is now *intr_countp[intr]. The indirection
is required to get a contiguous range of counters for vmstat
and so that the drivers depend more in the driver than on the
interrupt number (drivers could take turns using an interrupt
and the counts would remain correct). There is a separate
counter for each device and for each stray interrupt. In
1.1.5, stray interrupt 7 clobbers the count for device 7 or
something worse if there is no device 7 :-(.
# mask is now intr_mask[intr] (was already indirect).
o Entry points are now _XintrI and _XfastintrI (I = intr = 0-15),
not _VdevU (U = unit).
o Removed BUILD_VECTORS stuff. There's a trace of it left for
the string table for vmstat but config now generates the
string in one piece because nothing more is required.
o Removed old handling of stray interrupts and older comments
about it.
Submitted by: Bruce Evans
not provide the full accuracy of a randomized statistical clock, it does
provide greater accuracy than the previous method, while not significantly
increasing overhead. It also provides profiling support at 1024 Hz.
You must re-compile config before making a new kernel, or you will end
up with unresolved symbols.
Reviewed uy: Bruce evans said it worked for him.
Delete the ifdef GPL_EMULATE case here and made the padding work for
both types of emulators so that there is no longer a need to compile
ps and friends new if you are using the GPL math emulator instead the
normal one.
``changes'' are actually not changes at all, but CVS sometimes has trouble
telling the difference.
This also includes support for second-directory compiles. This is not
quite complete yet, as `config' doesn't yet do the right thing. You can
still make it work trivially, however, by doing the following:
rm /sys/compile
mkdir /usr/obj/sys/compile
ln -s M-. /sys/compile
cd /sys/i386/conf
config MYKERNEL
cd ../../compile/MYKERNEL
ln -s /sys @
rm machine
ln -s @/i386/include machine
make depend
make
improvements via the new routines pmap_qenter/pmap_qremove and pmap_kenter/
pmap_kremove. These routine allow fast mapping of pages for those
architectures that have "normal" MMUs. Also included is a fix to the
pageout daemon to properly check a queue end condition.
Submitted by: John Dyson
me:
1) TLB flush optimization that effectively eliminates half of all of the
TLB flushes. This works by only flushing the TLB when a page is "present"
in memory (i.e. the valid bit is set in the page table entry). See section
5.3.5 of the Intel 386 Programmer's Reference Manual.
2) The handling of "CMAP" has been improved to catch attempts at multiple
simultaneous use.
John:
1) Added pmap_qenter/pmap_qremove functions for fast mapping of pages into
the kernel. This is for future optimizations and support for the upcoming
merged VM/buffer cache.
Reviewed by: John Dyson
From Bruce Evans:
fu[i]byte() checked the wrong register. This caused interesting behaviour
in the GPL math emulator. The emulator does not check the values returned
by fu*() or su*() (:-() and it interpreted the address of -12(%ebp) as
-1(%ebp). The same probably occurs for all signed 8-bit offsets from
registers.
I cleaned up the new bzero() a bit.
Vastly improved trap.c from me. This rewritten version has a variety of
features, amoung them: higher performance and much higher code quality.
support.s, cpufunc.h:
No longer use gs override to enforce range limits - compare directly
against VM_MAXUSER_ADDRESS instead. The old way caused problems in
preserving the gs selector...and this method is just as fast or faster.
the NTP kernel PLL is disabled, and acquire_timer0() is enabled, thus
opening the door for microtime() (and hence gettimeofday()) to return
bogus timestamps. This option is necessary for the `pca' driver to
work, but is implemented to underscore the fact that accurate timekeeping
and the `pca' driver are incompatible at present. If someone writes a version
of microtime() that works when the `pca' driver is being used, this can get
junked.
1) check va before clearing the page clean flag. Not doing so was
causing the vnode pager error 5 messages when paging from
NFS. (pmap.c)
2) put back interrupt protection in idle_loop. Bruce didn't think
it was necessary, John insists that it is (and I agree). (swtch.s)
3) various improvements to the clustering code (vm_machdep.c). It's
now enabled/used by default.
4) bad disk blocks are now handled properly when doing clustered IOs.
(wd.c, vm_machdep.c)
5) bogus bad block handling fixed in wd.c.
6) algorithm improvements to the pageout/pagescan daemons. It's amazing
how well 4MB machines work now.
1) Removed all instances of disable_intr()/enable_intr() and changed
them back to splimp/splx. The previous method was done to improve
the performance, but Bruces recent changes to inline spl* have
made this unnecessary.
2) Cleaned up vm_machdep.c considerably. Probably fixed a few bugs, too.
3) Added a new mechanism for collecting page statistics - now done by
a new system process "pagescan". Previously this was done by the
pageout daemon, but this proved to be impractical.
4) Improved the page usage statistics gathering mechanism - performance is
much improved in small memory machines.
5) Modified mbuf.h to enable the support for an external free routine when
using mbuf clusters. Added appropriate glue in various places to
allow this to work.
6) Adapted a suggested change to the NFS code from Yuval Yurom to take
advantage of #5.
7) Added fault/swap statistics support.
1) fixed some bugs related to the bounce buffer code
2) vnode pager now supports clustered pageouts
3) experimental code for clustering all I/O via a new "cldisksort"
4) added >16MB check to Bustek driver
5) made some experimental algorithmic changes to the pageout daemon
6) fixed bugs in truncating mapped files (esp when mapped via NFS)
7) reorganized vnode pager I/O code
list of changes, I've made the following additional changes:
1) i386/include/ipl.h renamed to spl.h as the name conflicts with the
file of the same name in i386/isa/ipl.h.
2) changed all use of *mask (i.e. netmask, biomask, ttymask, etc) to
*_imask (net_imask, etc).
3) changed vestige of splnet use in if_is to splimp.
4) got rid of "impmask" completely (Bruce had gotten rid of netmask),
and are now using net_imask instead.
5) dozens of minor cruft to glue in Bruce's changes.
These require changes I made to config(8) as well, and thus it must
be rebuilt.
-DG
from Bruce Evans:
sio:
o No diff is supplied. Remove the define of setsofttty(). I hope
that is enough.
*.s:
o i386/isa/debug.h no longer exists. The event counters became too
much trouble to maintain. All function call entry and exception
entry counters can be recovered by using profiling kernel (the new
profiling supports all entry points; however, it is too slow to
leave enabled all the time; it also). Only BDBTRAP() from debug.h
is now used. That is moved to exception.s. It might be worth
preserving SHOW_BITS() and calling it from _mcount() (if enabled).
o T_ASTFLT is now only set just before calling trap().
o All exception handlers set SWI_AST_MASK in cpl as soon as possible
after entry and arrange for _doreti to restore it atomically with
exiting. It is not possible to set it atomically with entering
the kernel, so it must be checked against the user mode bits in
the trap frame before committing to using it. There is no place
to store the old value of cpl for syscalls or traps, so there are
some complications restoring it.
Profiling stuff (mostly in *.s):
o Changes to kern/subr_mcount.c, gcc and gprof are not supplied yet.
o All interesting labels `foo' are renamed `_foo' and all
uninteresting labels `_bar' are renamed `bar'. A small change
to gprof allows ignoring labels not starting with underscores.
o MCOUNT_LABEL() is to provide names for counters for times spent
in exception handlers.
o FAKE_MCOUNT() is a version of MCOUNT() suitable for exception
handlers. Its arg is the pc where the exception occurred. The
new mcount() pretends that this was a call from that pc to a
suitable MCOUNT_LABEL().
o MEXITCOUNT is to turn off any timer started by MCOUNT().
/usr/src/sys/i386/i386/exception.s:
o The non-BDB BPTTRAP() macros were doing a sti even when interrupts
were disabled when the trap occurred. The sti (fixed) sti is
actually a no-op unless you have my changes to machdep.c that make
the debugger trap gates interrupt gates, but fixing that would
make the ifdefs messier. ddb seems to be unharmed by both
interrupts always disabled and always enabled (I had the branch in
the fix back to front for some time :-().
o There is no known pushal bug.
o tf_err can be left as garbage for syscalls.
/usr/src/sys/i386/i386/locore.s:
o Fix and update BDE_DEBUGGER support.
o ENTRY(btext) before initialization was dangerous.
o Warm boot shot was longer than intended.
/usr/src/sys/i386/i386/machdep.c:
o DON'T APPLY ALL OF THIS DIFF. It's what I'm using, but may require
other changes.
Use the following:
o Remove aston() and setsoftclock().
Maybe use the following:
o No netisr.h.
o Spelling fix.
o Delay to read the Rebooting message.
o Fix for vm system unmapping a reduced area of memory
after bounds_check_with_label() reduces the size of
a physical i/o for a partition boundary. A similar
fix is required in kern_physio.c.
o Correct use of __CONCAT. It never worked here for non-
ANSI cpp's. Is it time to drop support for non-ANSI?
o gdt_segs init. 0xffffffffUL is bogus because ssd_limit
is not 32 bits. The replacement may have the same
value :-), but is more natural.
o physmem was one page too low. Confusing variable names.
Don't use the following:
o Better numbers of buffers. Each 8K page requires up to
16 buffer headers. On my system, this results in 5576
buffers containing [up to] 2854912 bytes of memory.
The usual allocation of about 384 buffers only holds
192K of disk if you use it on an fs with a block size
of 512.
o gdt changes for bdb.
o *TGT -> *IDT changes for bdb.
o #ifdefed changes for bdb.
/usr/src/sys/i386/i386/microtime.s:
o Use the correct asm macros. I think asm.h was copied from Mach
just for microtime and isn't used now. It certainly doesn't
belong in <sys>. Various macros are also duplicated in
sys/i386/boot.h and libc/i386/*.h.
o Don't switch to and from the IRR; it is guaranteed to be selected
(default after ICU init and explicitly selected in isa.c too, and
never changed until the old microtime clobbered it).
/usr/src/sys/i386/i386/support.s:
o Non-essential changes (none related to spls or profiling).
o Removed slow loads of %gs again. The LDT support may require
not relying on %gs, but loading it is not the way to fix it!
Some places (copyin ...) forgot to load it. Loading it clobbers
the user %gs. trap() still loads it after certain types of
faults so that fuword() etc can rely on it without loading it
explicitly. Exception handlers don't restore it. If we want
to preserve the user %gs, then the fastest method is to not
touch it except for context switches. Comparing with
VM_MAXUSER_ADDRESS and branching takes only 2 or 4 cycles on
a 486, while loading %gs takes 9 cycles and using it takes
another.
o Fixed a signed branch to unsigned.
/usr/src/sys/i386/i386/swtch.s:
o Move spl0() outside of idle loop.
o Remove cli/sti from idle loop. sw1 does a cli, and in the
unlikely event of an interrupt occurring and whichqs becoming
zero, sw1 will just jump back to _idle.
o There's no spl0() function in asm any more, so use splz().
o swtch() doesn't need to be superaligned, at least with the
new mcounting.
o Fixed a signed branch to unsigned.
o Removed astoff().
/usr/src/sys/i386/i386/trap.c:
o The decentralized extern decls were inconsistent, of course.
o Fixed typo MATH_EMULTATE in comments. */
o Removed unused variables.
o Old netmask is now impmask; print it instead. Perhaps we
should print some of the new masks.
o BTW, trap() should not print anything for normal debugger
traps.
/usr/src/sys/i386/include/asmacros.h:
o DON'T APPLY ALL OF THIS DIFF. Just use some of the null macros
as necessary.
/usr/src/sys/i386/include/cpu.h:
o CLKF_BASEPRI() changes since cpl == SWI_AST_MASK is now normal
while the kernel is running.
o Don't use var++ to set boolean variables. It fails after a mere
4G times :-) and is slower than storing a constant on [3-4]86s.
/usr/src/sys/i386/include/cpufunc.h:
o DON'T APPLY ALL OF THIS DIFF. You need mainly the include of
<machine/ipl.h>. Unfortunately, <machine/ipl.h> is needed by
almost everything for the inlines.
/usr/src/sys/i386/include/ipl.h:
o New file. Defines spl inlines and SWI macros and declares most
variables related to hard and soft interrupt masks.
/usr/src/sys/i386/isa/icu.h:
o Moved definitions to <machine/ipl.h>
/usr/src/sys/i386/isa/icu.s:
o Software interrupts (SWIs) and delayed hardware interrupts (HWIs)
are now handled uniformally, and dispatching them from splx() is
more like dispatching them from _doreti. The dispatcher is
essentially *(handler[ffs(ipending & ~cpl)]().
o More care (not quite enough) is taken to avoid unbounded nesting
of interrupts.
o The interface to softclock() is changed so that a trap frame is
not required.
o Fast interrupt handlers are now handled more uniformally.
Configuration is still too early (new handlers would require
bits in <machine/ipl.h> and functions to vector.s).
o splnnn() and splx() are no longer here; they are inline functions
(could be macros for other compilers). splz() is the nontrivial
part of the old splx().
/usr/src/sys/i386/isa/ipl.h
o New file. Supposed to have only bus-dependent stuff. Perhaps
the h/w masks should be declared here.
/usr/src/sys/i386/isa/isa.c:
o DON'T APPLY ALL OF THIS DIFF. You need only things involving
*mask and *MASK and comments about them. netmask is now a pure
software mask. It works like the softclock mask.
/usr/src/sys/i386/isa/vector.s:
o Reorganize AUTO_EOI* macros.
o Option FAST_INTR_HANDLER_USERS_ES for people who don't trust
fastintr handlers.
o fastintr handlers need to metamorphose into ordinary interrupt
handlers if their SWI bit has become set. Previously, sio had
unintended latency for handling output completions and input
of SLIP framing characters because this was not done.
/usr/src/sys/net/netisr.h:
o The machine-dependent stuff is now imported from <machine/ipl.h>.
/usr/src/sys/sys/systm.h
o DON'T APPLY ALL OF THIS DIFF. You need mainly the different
splx() prototype. The spl*() prototypes are duplicated as
inlines in <machine/ipl.h> but they need to be duplicated here
in case there are no inlines. I sent systm.h and cpufunc.h
to Garrett. We agree that spl0 should be replaced by splnone
and not the other way around like I've done.
/usr/src/sys/kern/kern_clock.c
o splsoftclock() now lowers cpl so the direct call to softclock()
works as intended.
o softclock() interface changed to avoid passing the whole frame
(some machines may need another change for profile_tick()).
o profiling renamed _profiling to avoid ANSI namespace pollution.
(I had to improve the mcount() interface and may as well fix it.)
The GUPROF variant doesn't actually reference profiling here,
but the 'U' in GUPROF should mean to select the microtimer
mcount() and not change the interface.
1) A new mechanism has been added to prevent pages from being paged
out called "vm_page_hold". Similar to vm_page_wire, but
much lower overhead.
2) Scheduling algorithm has been changed to improve interactive
performance.
3) Paging algorithm improved.
4) Some vnode and swap pager bugs fixed.
Eliminates vm_fault overhead on process startup and
mmap referenced data for in-memory pages.
(process startup time using in-memory segments *much* faster)
2) Even more efficient pmap code. Code partially cleaned up.
More comments yet to follow.
(generally more efficient pte management)
3) Pageout clustering ( in addition to the FreeBSD V1.1 pagein
clustering.)
(much faster paging performance on non-write behind disk
subsystems, slightly faster performance on other systems.)
4) Slightly changed vm_pageout code for more efficiency and
better statistics. Also, resist swapout a little more.
(less likely to pageout a recently used page)
5) Slight improvement to the page table page trap efficiency.
(generally faster system VM fault performance)
6) Defer creation of unnamed anonymous regions pager until needed.
(speeds up shared memory bss creation)
7) Remove possible deadlock from swap_pager initialization.
8) Enhanced procfs to provide "vminfo" about vm objects and user
pmaps.
9) Increased MCLSHIFT/MCLBYTES from 2K to 4K to improve net &
socket performance and to prepare for things to come.
John Dyson
dyson@implode.root.com
David Greenman
davidg@root.com
a pv chain was being traversed while interrupts were
fully enabled in pmap_remove_all ... this is bogus, and
has been fixed in pmap.c. (sorry for adding the splimp)
Description:
The integer overflow instruction (into) and the interrupt instruction with
value 4 (int #4) both give rise to SIGBUS signals rather than SIGFPE. The
problem is that overflow is a trap not a fault (unlike the BOUND instruction).
following is a summary:
1) increased object cache back up to a more reasonable value.
2) removed old & bogus cruft from machdep.c (clearseg, copyseg,
physcopyseg, etc).
3) inlined many functions in pmap.c
4) changed "load_cr3(rcr3())" into tlbflush() and made tlbflush inline
assembly.
5) changed the way that modified pages are tracked - now vm_page struct
is kept updated directly - no more scanning page tables.
6) removed lots of unnecessary spl's
7) removed old unused functions from pmap.c
8) removed all use of page_size, page_shift, page_mask variables - replaced
with PAGE_ constants.
9) moved trunc/round_page, atop, ptoa, out of vm_param.h and into i386/
include/param.h, and optimized them.
10) numerous changes to sys/vm/ swap_pager, vnode_pager, pageout, fault
code to improve performance. LRU algorithm modified to be more
effective, read ahead/behind values tuned for better performance,
etc, etc...
problem with some chipsets (UMC) remapping the 'hole' memory even when
you've got 16MB. People were led to believe that since there was only
16MB of memory in the machine, that they were okay wrt the ISA DMA
limit. This hack simply causes the extra memory to be ignored if it
appears around the 16MB limit.
