- Add racd* devices for MSCP based CD-ROM drives (RRD40s), as opposed
to mixing them in with the ra* disks, modelled loosely on how MSCP
rx* floppies are handled
- racd* defaults to the same single iso9660 partition as cd*
- Cleanup config option handling slightly so any combination of ra*,
rx* (and now raccd*) can be defined without build errors.
- Avoid ugly printf when disks have all zero disklabel magic number

branches: 1.7.2; 1.7.176; 1.7.182;
Bus'ify the mscp code also.

Copied from ../../arch/vax/mscp/files.mscp,v

Small Wall fixes.

Remove "class" declarations, and add "devclass" declarations where
appropriate. Fix several inconsistencies between device class and
attributes. Mostly from Chris Demetriou.

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Update for changes to config.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.

mscp(4): add ability for rronline() callback in a workqueue

when an ra(4) disk comes online the hardware interrupt ends up calling
disk_set_info(), which triggers a sleep lock/alloc in this path which
is triggered by LOCKDEBUG.

piggy-back on the existing workqueue for autoconfiguration to handle
this path and run the online completion in the work queue.

this is a little ugly, in that it puts two different types of work
into the one queue, but seems less ugly than creating a second
workqueue for what is likely another one-time event (infact, the other
user may be better handled via config_defer() -- i did not look too
closely.)

with this, LOCKDEBUG kernels work.

tested in simh.

branches: 1.38.6;
Merge thorpej-cfargs2.

branches: 1.37.8;
Merge thorpej-cfargs branch:

Simplify and make extensible the config_search() / config_found() /
config_attach() interfaces: rather than having different variants for
which arguments you want pass along, just have a single call that
takes a variadic list of tag-value arguments.

Adjust all call sites:
- Simplify wherever possible; don't pass along arguments that aren't
actually needed.
- Don't be explicit about what interface attribute is attaching if
the device only has one. (More simplification.)
- Add a config_probe() function to be used in indirect configuiration
situations, making is visibly easier to see when indirect config is
in play, and allowing for future change in semantics. (As of now,
this is just a wrapper around config_match(), but that is an
implementation detail.)

Remove unnecessary or redundant interface attributes where they're not
needed.

There are currently 5 "cfargs" defined:
- CFARG_SUBMATCH (submatch function for direct config)
- CFARG_SEARCH (search function for indirect config)
- CFARG_IATTR (interface attribte)
- CFARG_LOCATORS (locators array)
- CFARG_DEVHANDLE (devhandle_t - wraps OFW, ACPI, etc. handles)

...and a sentinel value CFARG_EOL.

Add some extra sanity checking to ensure that interface attributes
aren't ambiguous.

Use CFARG_DEVHANDLE in MI FDT, OFW, and ACPI code, and macppc and shark
ports to associate those device handles with device_t instance. This
will trickle trough to more places over time (need back-end for pre-OFW
Sun OBP; any others?).

branches: 1.36.54;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.35.2;
Adjust MSCP attach routines. The current code worked fine on simh-vax, but
unfortunately failed on at least one 'real' SCSI MSCP adaptor.
In the updated code mscp_attach() still iterates over each unit on a bus,
but mscp_dorsp() now detects if we have reached the last unit and does not
return a bogus additional unit. It also loses a few gotos and handles
noncontiguous unit numbers better.

branches: 1.34.12; 1.34.18;
struct device * -> device_t, no functional changes intended.

Remove extra whitespace added by a stupid tool.
XXX: more in src/sys/arch

bcopy -> memcpy

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

branches: 1.30.2;
mscp(4) was being a bad monkey! It's not OK to call config_found() from
interrupt context, as my vax kernel with DIAGNOSTIC told me. Instead, we
defer the work and get it to run in thread context via a workqueue(9).

Thanks to matt@ for the review.

branches: 1.29.4; 1.29.12;
use aprint_*_dev and device_xname

branches: 1.28.6;
B_BAD has done nothing for a long time.

branches: 1.27.2; 1.27.8;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.26.4; 1.26.6; 1.26.10; 1.26.12;
It's not a good idea for device drivers to modify b_flags, as they don't
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.

branches: 1.25.2; 1.25.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.24.14;
Use device_parent().

branches: 1.23.4; 1.23.6; 1.23.8; 1.23.10; 1.23.12;
merge ktrace-lwp.

branches: 1.22.4;
nuke trailing whitespace

branches: 1.21.4; 1.21.6;
Include the <sys/bufq.h> header to allow these to compile again.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.19.2;
Fix wrong argument to free due to malloc_type changes. While there,
spot blatant bug by inspection and fix that as well. (do not compare
scaled value against unscaled value).

Don't "extern int cold;" - this is in <sys/kernel.h>.

Call malloc(9) with M_ZERO flag instead of bzero() after malloc().

add RCSID

branches: 1.15.2;
Upon inspection of a 4000/600 dmesg, change the splbios to splubas.

XXX THIS CODE IS SERIOUSLY BROKEN! MSCP DEVICES ARE BLOCK I/O DEVICES,
AND THUS SPLBIO SHOULD BLOCK THEIR INTERRUPTS. The VAX port really
needs to get virtualized spl*() levels.

splimp -> splbio

branches: 1.13.6;
sleep() -> tsleep()

Remove register declarations.

branches: 1.11.2; 1.11.4;
Bus'ify the mscp code also.

Copied from ../../arch/vax/mscp/mscp.c,v

EGCS fixes + TU81 fix.

Small Wall fixes.

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Bugfix to mscp tape code; so that weof/eot works correctly.

Many minor bug fixes; mscp subsystem seems to work rather stable now.

backout previous kprintf change

printf -> kprintf, sprintf -> ksprintf

Add support for tape drives. More extensive error checking.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.

Adjust MSCP attach routines. The current code worked fine on simh-vax, but
unfortunately failed on at least one 'real' SCSI MSCP adaptor.
In the updated code mscp_attach() still iterates over each unit on a bus,
but mscp_dorsp() now detects if we have reached the last unit and does not
return a bogus additional unit. It also loses a few gotos and handles
noncontiguous unit numbers better.

branches: 1.8.110; 1.8.116;
merge ktrace-lwp.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.6.2; 1.6.38;
Copied from ../../arch/vax/mscp/mscp.h,v

Small Wall fixes.

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Bugfix to mscp tape code; so that weof/eot works correctly.

Add support for tape drives. More extensive error checking.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.

mscp(4): add ability for rronline() callback in a workqueue

when an ra(4) disk comes online the hardware interrupt ends up calling
disk_set_info(), which triggers a sleep lock/alloc in this path which
is triggered by LOCKDEBUG.

piggy-back on the existing workqueue for autoconfiguration to handle
this path and run the online completion in the work queue.

this is a little ugly, in that it puts two different types of work
into the one queue, but seems less ugly than creating a second
workqueue for what is likely another one-time event (infact, the other
user may be better handled via config_defer() -- i did not look too
closely.)

with this, LOCKDEBUG kernels work.

tested in simh.

branches: 1.90.4;
fix various typos in comments and messages.

branches: 1.89.2; 1.89.8;
Initialize disk_geom information, DIOCGPARTINFO relies on it.

Use C99-style initializers for struct dkdriver.

We have three sets of DTYPE_ constants in the kernel:
altq Drop Type
disklabel Disk Type
file Descriptor Type
(not to mention constants that contain the string DTYPE).
Let's make them two, by changing the disklabel one to be DisK TYPE since the
other disklabel constants seem to do that. Not many userland programs use
these constants (and the ones that they do are mostly in ifdefs). They will
be fixed shortly.

more unused removal

remove unused variable

make more drivers use disk_ioctl, and add a dev parameter to it so that
we can merge the "easy" disklabel ioctls to it. Ultimately all this will
go do dk_ioctl once all the drivers have been converted.

Centralize wedge ioctls in disk_ioctl.

branches: 1.82.2;
support DIOCMWEDGES ioctl

Add d_discard to all struct cdevsw instances I could find.

All have been set to "nodiscard"; some should get a real implementation.

Add d_discard to all struct bdevsw instances I could find.

I've set them all to nodiscard. Some of them (wd, dk, vnd, ld,
raidframe, maybe cgd) should be implemented for real.

branches: 1.79.2;
typo in previous

Change (mostly mechanically) every cdevsw/bdevsw I can find to use
designated initializers.

I have not built every extant kernel so I have probably broken at
least one build; however I've also found and fixed some wrong
cdevsw/bdevsw entries so even if so I think we come out ahead.

Eliminate unused variables

branches: 1.76.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.75.2;
- Unconditionally initialise tp to avoid a warning
- Use DK_CLOSED rather than CLOSED (no functional change)

revert unwanted part of previous MSCP commit

Adjust MSCP attach routines. The current code worked fine on simh-vax, but
unfortunately failed on at least one 'real' SCSI MSCP adaptor.
In the updated code
mscp_dorsp()

- Add racd* devices for MSCP based CD-ROM drives (RRD40s), as opposed
to mixing them in with the ra* disks, modelled loosely on how MSCP
rx* floppies are handled
- racd* defaults to the same single iso9660 partition as cd*
- Cleanup config option handling slightly so any combination of ra*,
rx* (and now raccd*) can be defined without build errors.
- Avoid ugly printf when disks have all zero disklabel magic number

branches: 1.71.12; 1.71.18;
Make this compile on ports with __HAVE_OLD_DISKLABEL
(though currently no such ports which have mscp).
Noticed by PR kern/41862.

struct device * -> device_t, no functional changes intended.

struct cfdata * -> cfdata_t, no functional changes intended.

Remove extra whitespace added by a stupid tool.
XXX: more in src/sys/arch

bcopy -> memcpy

bzero -> memset

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

branches: 1.63.2;
buildfix: re-adapt to major() return a 32bit value again.

g/c BUFQ_FOO() macros and use bufq_foo() directly.

use PRIu64 format to printf major() value

branches: 1.60.4;
use device_lookup_private() to get the softc -- while device/softc
are not split yet, it is better than a typecast

branches: 1.59.2; 1.59.4; 1.59.6; 1.59.8;
use aprint_*_dev and device_xname

branches: 1.58.16;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.57.2;
Merge disk init changes from the vmlocking branch. These seperate init /
destroy of 'struct disk' from attach / detach.

branches: 1.56.4; 1.56.6; 1.56.8; 1.56.10;
It's not a good idea for device drivers to modify b_flags, as they don't
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.

Replace some uses of lockmgr().

branches: 1.54.2; 1.54.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.53.14;
Use device_private().

Don't sleep on &rx->ra_dev.dv_unit -- sleep on &rx->ra_state instead.

Use device_unit().

Use device_parent().

branches: 1.49.4; 1.49.6; 1.49.8; 1.49.10; 1.49.12;
merge ktrace-lwp.

- change the way to specify a bufq strategy. (by string rather than by number)
- rather than embedding bufq_state in driver softc,
have a pointer to the former.
- move bufq related functions from kern/subr_disk.c to kern/subr_bufq.c.
- rename method to strategy for consistency.
- move some definitions which don't need to be exposed to the rest of kernel
from sys/bufq.h to sys/bufq_impl.h.
(is it better to move it to kern/ or somewhere?)
- fix some obvious breakage in dev/qbus/ts.c. (not tested)

branches: 1.47.4;
nuke trailing whitespace

de-__P

branches: 1.45.2; 1.45.4;
Make sure to get the right softc in rriodone.

Include the <sys/bufq.h> header to allow these to compile again.

Add wedges support to the MSCP disk driver.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.41.2;
Change bounds_check_with_label() to take a pointer to the disk structure,
rather than the label itself. This paves the way for some future changes.

Change return type of readdisklabel() to const char *
I hope I've found all the correct places!

Change SBSIZE to SBLOCKSIZE to make this compile after UFS2 was brought
in. OK'ed by ragge.

implement separate read/write disk statistics:
- disk_unbusy() gets a new parameter to tell the IO direction.
- struct disk_sysctl gets 4 new members for read/write bytes/transfers.
when processing hw.diskstats, add the read&write bytes/transfers for
the old combined stats to attempt to keep backwards compatibility.

unfortunately, due to multiple bugs, this will cause new kernels and old
vmstat/iostat/systat programs to fail. however, the next time this is
change it will not fail again.

this is just the kernel portion.

merge kqueue branch into -current

kqueue provides a stateful and efficient event notification framework
currently supported events include socket, file, directory, fifo,
pipe, tty and device changes, and monitoring of processes and signals

kqueue is supported by all writable filesystems in NetBSD tree
(with exception of Coda) and all device drivers supporting poll(2)

based on work done by Jonathan Lemon for FreeBSD
initial NetBSD port done by Luke Mewburn and Jason Thorpe

Add trailing ; to CFATTACH_DECL.

Use CFATTACH_DECL().

Declare all cfattach structures const.

Merge the gehenna-devsw branch into the trunk.

This merge changes the device switch tables from static array to
dynamically generated by config(8).

- All device switches is defined as a constant structure in device drivers.

- The new grammer ``device-major'' is introduced to ``files''.

device-major <prefix> char <num> [block <num>] [<rules>]

- All device major numbers must be listed up in port dependent majors.<arch>
by using this grammer.

- Added the new naming convention.
The name of the device switch must be <prefix>_[bc]devsw for auto-generation
of device switch tables.

- The backward compatibility of loading block/character device
switch by LKM framework is broken. This is necessary to convert
from block/character device major to device name in runtime and vice versa.

- The restriction to assign device major by LKM is completely removed.
We don't need to reserve LKM entries for dynamic loading of device switch.

- In compile time, device major numbers list is packed into the kernel and
the LKM framework will refer it to assign device major number dynamically.

Correct ra disk instrumentation. From Johnny Billquist.

Convert to new device buffer queue interface.

Approved by: Anders Magnusson <ragge@netbsd.org>

branches: 1.30.8; 1.30.10;
add RCSID

branches: 1.29.2; 1.29.4;
Upon inspection of a 4000/600 dmesg, change the splbios to splubas.

XXX THIS CODE IS SERIOUSLY BROKEN! MSCP DEVICES ARE BLOCK I/O DEVICES,
AND THUS SPLBIO SHOULD BLOCK THEIR INTERRUPTS. The VAX port really
needs to get virtualized spl*() levels.

splimp -> splbio

branches: 1.27.2;
Return error in the case of using ODIOCGDINFO or ODIOCGDEFLABEL when
the number of partitions is > OLDMAXPARTITIONS. This is better
than silently truncating the label (don't want to silently throw
away partitions when using an old disklabel binary on a label with
> 8 partitions). From Enami Tsugutomo.

Adapt all disk devices in MI directories to handle ODIOC* calls
for ports that have bumped MAXPARTITIONS (and thus define
__HAVE_OLD_DISKLABEL).

branches: 1.25.4;
sleep() -> tsleep()

Remove (more) VAX dependencies.

Remove register declarations.

Update for sys/buf.h/disksort_*() changes.

branches: 1.21.2; 1.21.4;
Bus'ify the mscp code also.

Copied from ../../arch/vax/mscp/mscp_disk.c,v

Fix disk size identifier bug.

Small Wall fixes.

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Remove __BROKEN_INDIRECT_CONFIG and change all drivers to use
config_search instead.

Update for changes to config.

use locator defines in "locators.h" to index cf_loc[]

foosize()'s return value is in DEV_BSIZE units; adjust the size obtained
from the disklabel accordingly.

Disk statistics fixes. From Johnny Billquist (bqt@update.uu.se).

Create a fake disklabel if there are no one on the disk.

Change the way boot device is figured out; now only using DEC numbers
and done when the device get attached.

ra disks is DV_DISK, not DV_DULL.

branches: 1.8.4;
Pass a struct device ** down to the functions that find the boot device.

branches: 1.7.2;
Many minor bug fixes; mscp subsystem seems to work rather stable now.

Use bitmask_snprintf().

backout previous kprintf change

printf -> kprintf, sprintf -> ksprintf

Automatic changing of root device after boot now works.

Add support for tape drives. More extensive error checking.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.

s/indicies/indices/ in comments.

fix various typos in comments.

remove checks for failure after memory allocation calls that cannot fail:

kmem_alloc() with KM_SLEEP
kmem_zalloc() with KM_SLEEP
percpu_alloc()
pserialize_create()
psref_class_create()

all of these paths include an assertion that the allocation has not failed,
so callers should not assert that again.

branches: 1.46.6;
Eliminate unused variables

branches: 1.45.4;
Fix device_t split glitch

split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.43.2;
Adjust MSCP attach routines. The current code worked fine on simh-vax, but
unfortunately failed on at least one 'real' SCSI MSCP adaptor.
In the updated code mscp_attach() still iterates over each unit on a bus,
but mscp_dorsp() now detects if we have reached the last unit and does not
return a bogus additional unit. It also loses a few gotos and handles
noncontiguous unit numbers better.

- Add racd* devices for MSCP based CD-ROM drives (RRD40s), as opposed
to mixing them in with the ra* disks, modelled loosely on how MSCP
rx* floppies are handled
- racd* defaults to the same single iso9660 partition as cd*
- Cleanup config option handling slightly so any combination of ra*,
rx* (and now raccd*) can be defined without build errors.
- Avoid ugly printf when disks have all zero disklabel magic number

branches: 1.41.12; 1.41.18;
struct device * -> device_t, no functional changes intended.

struct cfdata * -> cfdata_t, no functional changes intended.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

branches: 1.37.2;
mscp(4) was being a bad monkey! It's not OK to call config_found() from
interrupt context, as my vax kernel with DIAGNOSTIC told me. Instead, we
defer the work and get it to run in thread context via a workqueue(9).

Thanks to matt@ for the review.

g/c BUFQ_FOO() macros and use bufq_foo() directly.

branches: 1.35.4; 1.35.12;
use aprint_*_dev and device_xname

branches: 1.34.16;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.33.6; 1.33.8; 1.33.12;
fix typos

branches: 1.32.14; 1.32.18; 1.32.20;
Use device_private().

Use device_parent().

branches: 1.30.4; 1.30.6; 1.30.8; 1.30.10; 1.30.12;
merge ktrace-lwp.

- change the way to specify a bufq strategy. (by string rather than by number)
- rather than embedding bufq_state in driver softc,
have a pointer to the former.
- move bufq related functions from kern/subr_disk.c to kern/subr_bufq.c.
- rename method to strategy for consistency.
- move some definitions which don't need to be exposed to the rest of kernel
from sys/bufq.h to sys/bufq_impl.h.
(is it better to move it to kern/ or somewhere?)
- fix some obvious breakage in dev/qbus/ts.c. (not tested)

branches: 1.28.2;
Fixes due to cast-qual addition.

nuke trailing whitespace

de-__P

branches: 1.25.4; 1.25.6;
move buffer queue related stuffs from buf.h to their own header, bufq.h.

Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.23.2;
Use aprint_normal() in cfprint routines.

Add trailing ; to CFATTACH_DECL.

Use CFATTACH_DECL().

Declare all cfattach structures const.

Convert to new device buffer queue interface.

Approved by: Anders Magnusson <ragge@netbsd.org>

branches: 1.18.8;
add RCSID

branches: 1.17.2;
Upon inspection of a 4000/600 dmesg, change the splbios to splubas.

XXX THIS CODE IS SERIOUSLY BROKEN! MSCP DEVICES ARE BLOCK I/O DEVICES,
AND THUS SPLBIO SHOULD BLOCK THEIR INTERRUPTS. The VAX port really
needs to get virtualized spl*() levels.

splimp -> splbio

branches: 1.15.6;
Remove register declarations.

Allow 16 fragments for MSCP DMA.

Update for sys/buf.h/disksort_*() changes.

branches: 1.12.2; 1.12.4;
Bus'ify the mscp code also.

Copied from ../../arch/vax/mscp/mscp_subr.c,v

EGCS fixes + TU81 fix.

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Remove __BROKEN_INDIRECT_CONFIG and change all drivers to use
config_search instead.

Update for changes to config.

Many minor bug fixes; mscp subsystem seems to work rather stable now.

backout previous kprintf change

printf -> kprintf, sprintf -> ksprintf

Automatic changing of root device after boot now works.

Add support for tape drives. More extensive error checking.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.

mscp(4): add ability for rronline() callback in a workqueue

when an ra(4) disk comes online the hardware interrupt ends up calling
disk_set_info(), which triggers a sleep lock/alloc in this path which
is triggered by LOCKDEBUG.

piggy-back on the existing workqueue for autoconfiguration to handle
this path and run the online completion in the work queue.

this is a little ugly, in that it puts two different types of work
into the one queue, but seems less ugly than creating a second
workqueue for what is likely another one-time event (infact, the other
user may be better handled via config_defer() -- i did not look too
closely.)

with this, LOCKDEBUG kernels work.

tested in simh.

branches: 1.43.60;
Add d_discard to all struct cdevsw instances I could find.

All have been set to "nodiscard"; some should get a real implementation.

Add d_discard to all struct bdevsw instances I could find.

I've set them all to nodiscard. Some of them (wd, dk, vnd, ld,
raidframe, maybe cgd) should be implemented for real.

branches: 1.41.2;
Change (mostly mechanically) every cdevsw/bdevsw I can find to use
designated initializers.

I have not built every extant kernel so I have probably broken at
least one build; however I've also found and fixed some wrong
cdevsw/bdevsw entries so even if so I think we come out ahead.

Eliminate unused variables

branches: 1.39.2;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.38.12; 1.38.22;
struct device * -> device_t, no functional changes intended.

struct cfdata * -> cfdata_t, no functional changes intended.

ANSIfy another 1261 function definitions.
The only ones left in sys are beyond by sed script!
(or in sys/dist or sys/external)
Mostly they have function pointer parameters.

Change about 4500 of the K&R function definitions to ANSI ones.
There are still about 1600 left, but they have ',' or /* ... */
in the actual variable definitions - which my awk script doesn't handle.
There are also many that need () -> (void).
(The script does handle misordered arguments.)

branches: 1.34.4; 1.34.10;
use device_lookup_private() to get the softc -- while device/softc
are not split yet, it is better than a typecast

branches: 1.33.2; 1.33.4; 1.33.6; 1.33.8;
use aprint_*_dev and device_xname

branches: 1.32.16;
machine/{bus,cpu,intr}.h -> sys/{bus,cpu,intr}.h

branches: 1.31.4; 1.31.6; 1.31.10; 1.31.12;
It's not a good idea for device drivers to modify b_flags, as they don't
need to understand the locking around that field. Instead of setting
B_ERROR, set b_error instead. b_error is 'owned' by whoever completes
the I/O request.

branches: 1.30.2; 1.30.10;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.29.14;
Use device_private().

Use device_parent().

branches: 1.27.2; 1.27.4;
Don't cast lvalue. remove uninitialized reference.

branches: 1.26.4; 1.26.6; 1.26.8;
merge ktrace-lwp.

branches: 1.25.2;
Fixes due to cast-qual addition.

nuke trailing whitespace

de-__P

branches: 1.22.4; 1.22.6;
Include the <sys/bufq.h> header to allow these to compile again.

branches: 1.21.6;
merge kqueue branch into -current

kqueue provides a stateful and efficient event notification framework
currently supported events include socket, file, directory, fifo,
pipe, tty and device changes, and monitoring of processes and signals

kqueue is supported by all writable filesystems in NetBSD tree
(with exception of Coda) and all device drivers supporting poll(2)

based on work done by Jonathan Lemon for FreeBSD
initial NetBSD port done by Luke Mewburn and Jason Thorpe

Add trailing ; to CFATTACH_DECL.

Use CFATTACH_DECL().

Declare all cfattach structures const.

Merge the gehenna-devsw branch into the trunk.

This merge changes the device switch tables from static array to
dynamically generated by config(8).

- All device switches is defined as a constant structure in device drivers.

- The new grammer ``device-major'' is introduced to ``files''.

device-major <prefix> char <num> [block <num>] [<rules>]

- All device major numbers must be listed up in port dependent majors.<arch>
by using this grammer.

- Added the new naming convention.
The name of the device switch must be <prefix>_[bc]devsw for auto-generation
of device switch tables.

- The backward compatibility of loading block/character device
switch by LKM framework is broken. This is necessary to convert
from block/character device major to device name in runtime and vice versa.

- The restriction to assign device major by LKM is completely removed.
We don't need to reserve LKM entries for dynamic loading of device switch.

- In compile time, device major numbers list is packed into the kernel and
the LKM framework will refer it to assign device major number dynamically.

branches: 1.16.8;
add RCSID

branches: 1.15.6; 1.15.8; 1.15.10;
Remove register declarations.

branches: 1.14.2; 1.14.4;
Bus'ify the mscp code also.

Copied from ../../arch/vax/mscp/mscp_tape.c,v

Fix deadlock bug. Pointed out by maximum entropy.

EGCS fixes + TU81 fix.

Small Wall fixes.

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Remove __BROKEN_INDIRECT_CONFIG and change all drivers to use
config_search instead.

Update for changes to config.

use locator defines in "locators.h" to index cf_loc[]

Bugfix to mscp tape code; so that weof/eot works correctly.

Many minor bug fixes; mscp subsystem seems to work rather stable now.

backout previous kprintf change

printf -> kprintf, sprintf -> ksprintf

Add support for tape drives. More extensive error checking.

merge ktrace-lwp.

nuke trailing whitespace

branches: 1.5.8; 1.5.10;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.4.2; 1.4.38;
Copied from ../../arch/vax/mscp/mscpreg.h,v

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Make boot compile again.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.

mscp(4): add ability for rronline() callback in a workqueue

when an ra(4) disk comes online the hardware interrupt ends up calling
disk_set_info(), which triggers a sleep lock/alloc in this path which
is triggered by LOCKDEBUG.

piggy-back on the existing workqueue for autoconfiguration to handle
this path and run the online completion in the work queue.

this is a little ugly, in that it puts two different types of work
into the one queue, but seems less ugly than creating a second
workqueue for what is likely another one-time event (infact, the other
user may be better handled via config_defer() -- i did not look too
closely.)

with this, LOCKDEBUG kernels work.

tested in simh.

branches: 1.18.70;
split device_t/softc for all remaining drivers.
replace "struct device *" with "device_t".
use device_xname(), device_unit(), etc.

branches: 1.17.12; 1.17.22;
struct device * -> device_t, no functional changes intended.

branches: 1.16.2;
mscp(4) was being a bad monkey! It's not OK to call config_found() from
interrupt context, as my vax kernel with DIAGNOSTIC told me. Instead, we
defer the work and get it to run in thread context via a workqueue(9).

Thanks to matt@ for the review.

branches: 1.15.40; 1.15.50;
Kill caddr_t; there will be some MI fallout, but it will be fixed shortly.

branches: 1.14.26;
merge ktrace-lwp.

- change the way to specify a bufq strategy. (by string rather than by number)
- rather than embedding bufq_state in driver softc,
have a pointer to the former.
- move bufq related functions from kern/subr_disk.c to kern/subr_bufq.c.
- rename method to strategy for consistency.
- move some definitions which don't need to be exposed to the rest of kernel
from sys/bufq.h to sys/bufq_impl.h.
(is it better to move it to kern/ or somewhere?)
- fix some obvious breakage in dev/qbus/ts.c. (not tested)

branches: 1.12.2;
Fixes due to cast-qual addition.

de-__P

branches: 1.10.8; 1.10.10;
Move UCB-licensed code from 4-clause to 3-clause licence.

Patches provided by Joel Baker in PR 22364, verified by myself.

branches: 1.9.6;
Convert to new device buffer queue interface.

Approved by: Anders Magnusson <ragge@netbsd.org>

branches: 1.8.6; 1.8.8; 1.8.20;
Update for sys/buf.h/disksort_*() changes.

branches: 1.7.2; 1.7.4;
Bus'ify the mscp code also.

Copied from ../../arch/vax/mscp/mscpvar.h,v

Split out MSCP floppies as a separate device (rx) instead of handling
them as disk drives. It is now also allowed to change floppies without
the system complaining :-)

Many minor bug fixes; mscp subsystem seems to work rather stable now.

Automatic changing of root device after boot now works.

Add support for tape drives. More extensive error checking.

Totally new approach for MSCP devices; they are now handled more like SCSI.
All Unibus dependencies are removed.
Can support different controllers on different buses.
Allows cloning of devices.
TODO:
Write support for MSCP tapes.