fix double words in comments
feedback and ok jmc@ miod, ok millert@

stp(4): tsleep(9) -> tsleep_nsec(9)

With input from kn@.

Compile-tested by kn@.

ok kn@

Convert infinite sleeps to {m,t}sleep_nsec(9).

ok kn@

Convert K&R function definitions to modern C. clang is more picky about
them.
OK otto@ deraadt@

Reduces the noise around the global ``ticks'' variable by renaming
all the local ones to ``nticks''.

ok stefan@, deraadt@

oops

simplify kthread_create(). no more stdarg
ok matthew guenther mikeb

Use soft interrupts to dispatch pcmcia device interrupts; this ensures the
driver handlers get invoked at the right level. Parts from NetBSD.

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code that NetBSD has claimed for itself without attributing to the
actual authors.

OK deraadt@

use proper bus_size_t for register offsets and not bus_addr_t; uwe@ miod@ testing

Add pcmcia_intr_string() to format the intr string for pcmcia attach
routines.

The socket number diagnostic check was simply crying out for unsigned
arithmetic...

Rework the card insertion/removal detection logic:
- instead of queueing events each time the sense information changes,
only remember that the slot needs attention
- when processing a slot needing attention, decide what to do depending upon
its current sense information, rather than what it was when the change
was detected
- handle the ``only one sensor asserted'' state happening during slow
insertion or removal.

This makes the driver much more reactive to insertions and removals, and
also removes unnecessary attach/detach calls.

ok deraadt@

fix #if/ifdef issue and use u_int16_t instead of int where appropriate

Acknowledge the status interrupt in some odd and usually rare cases, which
could render a sparc so crazy it would panic...

mimic delays pattern from i82365 on socket enable. actually pull the reset line before clearing it. this together makes some more cards work and unwedges my wedged wedgy on reboot alright; miod@ ok and testing

fix all the _BITS and their use

Let the stp driver use different bus_space_tag_t for its own device access,
and for access to the pcmcia windows from the pcmcia drivers.

Split the stp driver into its core, and md attachment routines, and provide
attachment routines for sparc and sparc64.

branches: 1.4.4; 1.4.8;
make this more quiet on card insertion.

rename nell to stp. we do not always follow stupid solaris names

- set the size correctly.
- a few minor cosmetical changes.

sbus pcmcia bridge driver (nell); from NetBSD.

stp(4): tsleep(9) -> tsleep_nsec(9)

With input from kn@.

Compile-tested by kn@.

ok kn@

Convert infinite sleeps to {m,t}sleep_nsec(9).

ok kn@

Convert K&R function definitions to modern C. clang is more picky about
them.
OK otto@ deraadt@

Reduces the noise around the global ``ticks'' variable by renaming
all the local ones to ``nticks''.

ok stefan@, deraadt@

oops

simplify kthread_create(). no more stdarg
ok matthew guenther mikeb

Use soft interrupts to dispatch pcmcia device interrupts; this ensures the
driver handlers get invoked at the right level. Parts from NetBSD.

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code that NetBSD has claimed for itself without attributing to the
actual authors.

OK deraadt@

use proper bus_size_t for register offsets and not bus_addr_t; uwe@ miod@ testing

Add pcmcia_intr_string() to format the intr string for pcmcia attach
routines.

The socket number diagnostic check was simply crying out for unsigned
arithmetic...

Rework the card insertion/removal detection logic:
- instead of queueing events each time the sense information changes,
only remember that the slot needs attention
- when processing a slot needing attention, decide what to do depending upon
its current sense information, rather than what it was when the change
was detected
- handle the ``only one sensor asserted'' state happening during slow
insertion or removal.

This makes the driver much more reactive to insertions and removals, and
also removes unnecessary attach/detach calls.

ok deraadt@

fix #if/ifdef issue and use u_int16_t instead of int where appropriate

Acknowledge the status interrupt in some odd and usually rare cases, which
could render a sparc so crazy it would panic...

mimic delays pattern from i82365 on socket enable. actually pull the reset line before clearing it. this together makes some more cards work and unwedges my wedged wedgy on reboot alright; miod@ ok and testing

fix all the _BITS and their use

Let the stp driver use different bus_space_tag_t for its own device access,
and for access to the pcmcia windows from the pcmcia drivers.

Split the stp driver into its core, and md attachment routines, and provide
attachment routines for sparc and sparc64.

branches: 1.4.4; 1.4.8;
make this more quiet on card insertion.

rename nell to stp. we do not always follow stupid solaris names

- set the size correctly.
- a few minor cosmetical changes.

sbus pcmcia bridge driver (nell); from NetBSD.

Convert infinite sleeps to {m,t}sleep_nsec(9).

ok kn@

Convert K&R function definitions to modern C. clang is more picky about
them.
OK otto@ deraadt@

Reduces the noise around the global ``ticks'' variable by renaming
all the local ones to ``nticks''.

ok stefan@, deraadt@

oops

simplify kthread_create(). no more stdarg
ok matthew guenther mikeb

Use soft interrupts to dispatch pcmcia device interrupts; this ensures the
driver handlers get invoked at the right level. Parts from NetBSD.

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code that NetBSD has claimed for itself without attributing to the
actual authors.

OK deraadt@

use proper bus_size_t for register offsets and not bus_addr_t; uwe@ miod@ testing

Add pcmcia_intr_string() to format the intr string for pcmcia attach
routines.

The socket number diagnostic check was simply crying out for unsigned
arithmetic...

Rework the card insertion/removal detection logic:
- instead of queueing events each time the sense information changes,
only remember that the slot needs attention
- when processing a slot needing attention, decide what to do depending upon
its current sense information, rather than what it was when the change
was detected
- handle the ``only one sensor asserted'' state happening during slow
insertion or removal.

This makes the driver much more reactive to insertions and removals, and
also removes unnecessary attach/detach calls.

ok deraadt@

fix #if/ifdef issue and use u_int16_t instead of int where appropriate

Acknowledge the status interrupt in some odd and usually rare cases, which
could render a sparc so crazy it would panic...

mimic delays pattern from i82365 on socket enable. actually pull the reset line before clearing it. this together makes some more cards work and unwedges my wedged wedgy on reboot alright; miod@ ok and testing

fix all the _BITS and their use

Let the stp driver use different bus_space_tag_t for its own device access,
and for access to the pcmcia windows from the pcmcia drivers.

Split the stp driver into its core, and md attachment routines, and provide
attachment routines for sparc and sparc64.

branches: 1.4.4; 1.4.8;
make this more quiet on card insertion.

rename nell to stp. we do not always follow stupid solaris names

- set the size correctly.
- a few minor cosmetical changes.

sbus pcmcia bridge driver (nell); from NetBSD.

Convert K&R function definitions to modern C. clang is more picky about
them.
OK otto@ deraadt@

Reduces the noise around the global ``ticks'' variable by renaming
all the local ones to ``nticks''.

ok stefan@, deraadt@

oops

simplify kthread_create(). no more stdarg
ok matthew guenther mikeb

Use soft interrupts to dispatch pcmcia device interrupts; this ensures the
driver handlers get invoked at the right level. Parts from NetBSD.

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code that NetBSD has claimed for itself without attributing to the
actual authors.

OK deraadt@

use proper bus_size_t for register offsets and not bus_addr_t; uwe@ miod@ testing

Add pcmcia_intr_string() to format the intr string for pcmcia attach
routines.

The socket number diagnostic check was simply crying out for unsigned
arithmetic...

Rework the card insertion/removal detection logic:
- instead of queueing events each time the sense information changes,
only remember that the slot needs attention
- when processing a slot needing attention, decide what to do depending upon
its current sense information, rather than what it was when the change
was detected
- handle the ``only one sensor asserted'' state happening during slow
insertion or removal.

This makes the driver much more reactive to insertions and removals, and
also removes unnecessary attach/detach calls.

ok deraadt@

fix #if/ifdef issue and use u_int16_t instead of int where appropriate

Acknowledge the status interrupt in some odd and usually rare cases, which
could render a sparc so crazy it would panic...

mimic delays pattern from i82365 on socket enable. actually pull the reset line before clearing it. this together makes some more cards work and unwedges my wedged wedgy on reboot alright; miod@ ok and testing

fix all the _BITS and their use

Let the stp driver use different bus_space_tag_t for its own device access,
and for access to the pcmcia windows from the pcmcia drivers.

Split the stp driver into its core, and md attachment routines, and provide
attachment routines for sparc and sparc64.

branches: 1.4.4; 1.4.8;
make this more quiet on card insertion.

rename nell to stp. we do not always follow stupid solaris names

- set the size correctly.
- a few minor cosmetical changes.

sbus pcmcia bridge driver (nell); from NetBSD.

Reduces the noise around the global ``ticks'' variable by renaming
all the local ones to ``nticks''.

ok stefan@, deraadt@

oops

simplify kthread_create(). no more stdarg
ok matthew guenther mikeb

Use soft interrupts to dispatch pcmcia device interrupts; this ensures the
driver handlers get invoked at the right level. Parts from NetBSD.

First pass at removing clauses 3 and 4 from NetBSD licenses.

Not sure what's more surprising: how long it took for NetBSD to
catch up to the rest of the BSDs (including UCB), or the amount of
code that NetBSD has claimed for itself without attributing to the
actual authors.

OK deraadt@

use proper bus_size_t for register offsets and not bus_addr_t; uwe@ miod@ testing

Add pcmcia_intr_string() to format the intr string for pcmcia attach
routines.

The socket number diagnostic check was simply crying out for unsigned
arithmetic...

Rework the card insertion/removal detection logic:
- instead of queueing events each time the sense information changes,
only remember that the slot needs attention
- when processing a slot needing attention, decide what to do depending upon
its current sense information, rather than what it was when the change
was detected
- handle the ``only one sensor asserted'' state happening during slow
insertion or removal.

This makes the driver much more reactive to insertions and removals, and
also removes unnecessary attach/detach calls.

ok deraadt@

fix #if/ifdef issue and use u_int16_t instead of int where appropriate

Acknowledge the status interrupt in some odd and usually rare cases, which
could render a sparc so crazy it would panic...

mimic delays pattern from i82365 on socket enable. actually pull the reset line before clearing it. this together makes some more cards work and unwedges my wedged wedgy on reboot alright; miod@ ok and testing

fix all the _BITS and their use

Let the stp driver use different bus_space_tag_t for its own device access,
and for access to the pcmcia windows from the pcmcia drivers.

Split the stp driver into its core, and md attachment routines, and provide
attachment routines for sparc and sparc64.

branches: 1.4.4; 1.4.8;
make this more quiet on card insertion.

rename nell to stp. we do not always follow stupid solaris names

- set the size correctly.
- a few minor cosmetical changes.

sbus pcmcia bridge driver (nell); from NetBSD.