| Age | Commit message (Collapse) | Author |
|---|
|
this gives us working glass console on Fuel, as well as on Octane systems
with Odyssey graphics.
Joint work with jsing@
|
|
fake ARCBios component structures associated to the KL configuration.
The ARCBios data tells us if the device is the output console, and the
KL component data tells us its node and widget numbers.
|
|
number the PCI bus they are on is connected to. Will be used shortly to help
the console device selection logic.
|
|
ok deraadt@
|
|
|
|
like bigmem (disabled by default), sppp (ask wilfried about
"splassert: if_up/down: want 5 have 7")
|
|
|
|
installboot on a mounted partition;
from J.C. Roberts
|
|
|
|
card, which can be found in Octane, Octane2, Fuel, Tezro and Onyx systems.
ok miod@
|
|
|
|
|
|
Much prodding from deraadt.
|
|
about, and never implemented in sparc64 but changes are the same.
ok deraadt@
|
|
|
|
uvm_km_alloc() (which allocates some virtual space and then
$size_of_mapping pages which it then enters) *then* pmap_kenter the
bus_space address that we wish to map, it's just a little bit stupid and
a waste of memory.
replace with uvm_km_valloc and ponies for everyone.
ok drahn@
|
|
ok miod@ otto@
|
|
make console selection on a per machine basis. Whilst here store the
keyboard layout ('keybd') and graphics state ('gfx') variables for
future use.
ok miod@
|
|
the serial and IR ports, built into a mac-like case). At least PMON initializes
the frame buffer in a much friendlier video mode (1360x768x16), but there is
still no frame buffer acceleration yet.
Tested by wvdputte.
|
|
attempts to map resources without bothering to know their size. I should
probably be more careful and do the BAR dance to get the BAR size, but then
at this point we are reusing mappings set up by PMON, and it's ok to trust it.
This would only have ever become an issue with a framebuffer larger than 64MB
mapped at PCIHI with a PCILO mapping overlapping the first few 64MB anyway.
|
|
info; ok miod@
|
|
condition to determine crtical status for the sensor and apm. This makes
the critical status correspond to the flashing red battery led.
Furthermore, use ISSET macros and friends.
|
|
- properly spell 'usefulness'
|
|
for the interrupt. Makes the vmstat -i output similar to what we see on
i386.
ok deraadt@, krw@
|
|
|
|
handlers now checks that the spl the isr was established at is the same as
the one passed to splassert. this lets you check that isrs dont enter code
that have insufficient protection if entered from process context.
ok kettenis@
|
|
and get interrupts by otto@, no further testing due to lack of IR devices.
|
|
Currently unable to change video modes or provide any form of acceleration,
so you are stuck in a 640x400x8 mode, but at least people scared of serial
consoles will get a chance to use their Fuloongs now.
Tested by otto@ (early developments) and jasper@ (final version)
|
|
longers assumes all requested mappings fit in the three 64MB PCILO windows,
but will instead check whether the requested mappings can be provided by
PCILO regions or PCIHI regions (and on 2F-based systems, making sure we
only use addresses which get properly routed from CPU to PCI within the
2F crossbar).
This in turn requires early console code to abide the bus_space rules and
get its resources from bus_space_map() rather than doing PCI BAR arithmetic
by itself.
No functional change on Lemote Yeeloong and Gdium Liberty; on Lemote Fuloong
2F this allows BAR set up in PCIHI space by PMON to be mapped by kernel code
without having to compensate for the PCILO offset.
|
|
|
|
ok marco deraadt
|
|
processors can display correct data. Now cpu1 on octane is correctly
reported in dmesg.
|
|
|
|
be decoupled from the nominal processor speed.
While there, make sure delay() gets a proper delay constant if invoked before
cpu0 attaches (how could I miss that when introducing struct cpu_hwinfo?!?)
|
|
|
|
devices. Original code was inspired by/copied/stolen from the macppc
apm(4), which has different semantics for battery life while charging.
|
|
"we dont' know"
|
|
generic part, ykbec(4) provides the device specific parts. Other
battery/power status drivers can easily hook to adb(4). With help
from miod@; ok matthieu@ miod@ jasper@
|
|
anyway, it's not worth fixing. Bump version number.
ok deraadt@
|
|
buffer cache first, so that all sorts of gibble doesn't end up on the disk.
ok kettenis (for the sparc/sparc64 ones, at least)
|
|
|
|
|
|
just not there.
ok jakemsr@ drahn@
|
|
Why this code, which must be executed, is not failing on disk is an utter
mystery. More investigation needed.
|
|
allowed to match on old gdium artwork, as newer artwork have nothing
connected to the usb pins, and this causes the empty bus probe to stall the
boot a few seconds.
|
|
but documentation for it only exists in Chinese. I still don't have the
slightest idea why flipping it is important, but I am more confident flipping
it now (-:
|
|
information and a few other things not yet wired.
|
|
|
|
No functional change.
|
|
|
