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.
in the future, these drivers won't need to maintain an array of
configured units. They still need to because ISA interrupt handlers
take a unit number. :(
Pass "softc" pointers instead of unit numbers to many functions that
did a conversion of unit->softc anyway.
this driver no longer needs to maintain an array of configured units.
Pass "softc" pointers instead of unit numbers to many functions that
did a conversion for unit->softc anyway.
whether of not to automatically #define EXCLUDE_AUDIO; MSS is a real
audio device and we should not #define EXCLUDE_AUDIO if we have one.
(And I want it because it's the only mixer-capable audio driver that I
can use with my crummy Packard Bell (nee Aztech) audio board.)
This fixes the very confusing condition where having all of this:
mss0 at 0x530 irq 10 drq 1 on isa
gus0: <MS Sound System (CS4231)>
opl0 at 0x388 on isa
opl0: <Yamaha OPL-3 FM>
mpu0 at 0x300 irq 9 drq 0 on isa
mpu0: <MPU-401 MIDI Interface 0.0 >
will still give you this:
% cat /dev/sndstat
SoundCard Error: The soundcard system has not been configured
Also remove an unnecessary newline in the printf() message for the
'gus0' device shown above so that we don't wind up printing a blank
line between mss0 and gus0.
was overlapping with another file, and making some undesirable behavior a
little worse - it's triggering a bug in config that appears to have been
there for some time (before the options files, anyway.)
to enable IP forwarding, use sysctl(8). Also did the same for IPX,
which involved inventing a completely new MIB from whole cloth (which
I may not quite have correct); be aware of this if you use IPX forwarding.
(The two should never have been controlled by the same option anyway.)
Simplify the initialization of adapters by pulling all card specific
initialization to the card specific modules.
Update comments and fix formating.
Pass struct ahc_data*'s to functions instead of unit numbers.
Take advantage of the quad word alignment of SCB fields.
Adapt to new sequencer changes:
1) Waiting scb list no longer has a tail.
2) Fill the message buffer as appropriate during a parity error.
3) Count all of the SGs involved in a residual instead of just
the current one.
The reset/abort code still needs a lot of work.
Reviewed by: David Greenman <davidg@FreeBSd.org>
aic7770.c:
Simplify the initialization of adapters by pulling all card specific
initialization to the card specific modules.
eisaconf.c:
outb 0x80 instead of 0xc80. The top byte is truncated anyway, and 0x80
was what was intended.
enough nodes for the number of ports on the last module, not the number
of ports _total_ that the driver is managing...
Submitted by: Robert Sanders <rsanders@mindspring.com>
Nobody in our regular source tree, or in the non-distfile part of the
ports tree does use /dev/io anyway, so this might be replaced by
another scenario some day.
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.
we can see if it's a small distance beyond the end, or way out. This may
give some clues as to whether it is being caused by something coalescing
the transfers in spite of the bounce buffers, or simply because of buffer
corruption. (The BT driver seems to occasionally get hit by from this too,
except that it does not trap the transfer, and the system panics later
with vm_bounce_page_free.) This "event" usually happens to me during a
savecore (on the rare occasion that a kernel coredump is actually taken
after a crash - the lack of kernel core dumps is another problem...).
bzero.
Deprecated blkclr (removed it).
Removed some old cruft from cpufunc.h.
The optimized bzero was submitted by Torbjorn Granlund <tege@matematik.su.se>
The kernel adaption and other changes by me.
bzero.
Deprecated blkclr (removed it).
Removed some old cruft from cpufunc.h.
The optimized bzero was submitted by Torbjorn Granlund <tege@matematik.su.se>
The kernel adaption and other changes by me.
bcopy:
Be smarter about handling overlapped copies and only go backwards if it
is really necessary. Going backwards on a P6 is much slower than forwards
and it's a little slower on a P5. Also moved the count mask and 'std'
down a few lines - it's a couple percent faster this way on a P5.
libkern.a are now specified by listing their source files in
files.${MACHINE}. The list is machine-dependent to save space.
All the necessary object for each machine must be linked into the
kernel in case an lkm wants one.
feature of the ICU. auto-EOI on the slave is not safe, however, so it
remains an option. Killed religious FASTER_NOP when writing the ICU.
Reviewed by: bde
or deleted.
Motivated by: `int doclusteread = 1;' in ext2_vnops.c redefined
doclusterread if DEBUG is defined, so it could not have worked.
This was fixed by staticizing things before it caused problems.
I didn't find any more cases like this.
wasteful, but better than clobbering the variables below the stack.
About 300 bytes of variables were clobbered when I examined double
faults using ddb. Perhaps a page that is known not to be accessed by
the double fault handler could be used. Such pages are not easy to
find, since the double fault handler calls panic() which calls sync()
and possibly dumpsys().
part of the DMA channel 0 address and wasn't random in the intended
way.
Submitted by: KATO Takenori <kato@eclogite.eps.nagoya-u.ac.jp>
Disable interrupts while reading the clock. This probably isn't
important (allowing interrupts probably increased randomness in
the usual case).
Removed __i386__ ifdef. This file is in an i386 directory and has
other i386 dependencies.
unintentionally committed):
- the fifo was completely disabled for low speeds. Apart from being
unnecessarily inefficient, this invalidated com->tx_fifo_size.
- `ftl' became a bogus name.
- the 16650 probe breaks the COM_NOFIFO() case and has other bugs
(disabled, not fixed).
Fixed bogus change of the fifo settings for the non-speed of 0. This
bug made the above fifo bug occur even at non-low speeds.
Fixed the modes of the cua devices. It isn't possible to set the uid
and gid correctly since the kernel can't know who uucp.dialer is.
Register the devswitch at device attach time. SYSINIT() is not
the right way to initialize devswitches (if anything :->).
Eventually, the devswitch should be deregistered at device detach
and/or unload time and reregistered at device attach time ... Then
some com->gone tests could be removed.
Cleaned up some other recent changes.
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.
about decoding trap/syscall/interrupt frames and generally works better
than the previous stuff.
Removed some special (incorrect) frobbing of the frame pointer that
was messing some things up with the new traceback code.
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.
prototypes don't go missing again. Also added -Winline so that some
doubtful (non-)inlines get fixed.
bsd.kmod.mk:
Also added `-Wreturn-type -Wimplicit -Wnested-externs' to catch up
with the kernel.
#includes to get prototypes.
pci now uses a different interrupt handler type for interrupts that it
dispatches and the isa interrupt handler type for the interrupts that
it handles.
successfully run linux netscape 2.0b3 with a QMAGIC ld.so and libc/libm
that I found on some linux machine that I _think_ is running slackware 3.0.
There are still problems.. ld.so claims the libraries are the wrong
format, but it still runs anyway.. :-/ The QMAGIC ld.so also screams
about needing ld.so.cache, and running a linux ldconfig is quite
educational. You soon learn to run "chroot /compat/linux /bin/ldconfig"
where ldconfig is living in /compat/linux/bin. :-]
(Lets just say that it puts loads of symlinks in /usr/lib otherwise :-)
