- Reduce available SPL levels for hardware devices to none, vm, sched, high.
- Acquire kernel_lock only for interrupts at IPL_VM.
- Implement threaded soft interrupts.
SEMMNI, SEMMNS, SEMUME and SHMMAXPGS.
They can be tweaked via sysctl now. Ports that were setting values on
them weren't touched, I only removed the ones that were commented out.
This branch was a major cleanup and rototill of many of the various OEA
cpu based PPC ports that focused on sharing as much code as possible
between the various ports to eliminate near-identical copies of files in
every tree. Additionally there is a new PIC system that unifies the
interface to interrupt code for all different OEA ppc arches. The work
for this branch was done by a variety of people, too long to list here.
TODO:
bebox still needs work to complete the transition to -renovation.
ofppc still needs a bunch of work, which I will be looking at.
ev64260 still needs to be renovated
amigappc was not attempted.
NOTES:
pmppc was removed as an arch, and moved to a evbppc target.
based machine. When we do, we have to frob a few special bits on the
pcib to finish the setup that the prom didn't bother completing for us.
With this change, NetBSD finally runs on the MTX604 board.. Many thanks
to John Baker for providing remote access to the machine, and rebooting
it repeatedly for me.
on the pin, to get us the right pin on the underlying PReP bridge to wire
the interrupt to. Tested with a 4-port fxp card on a 7043-140. Thanks
to simon for helping me get the math right.
associated with this, 0, the nvram device, and now 1, the residual
device. The devices allow the user to directly read the contents of the
kernel copy of the nvram, and the residual data respectively. There is
no provision for write.
actually contains entries. It is apparently possible to get a non-null,
but empty dictionary back. This seems to happen if you inject a ppb-card
on pci0 when you also have a prep pci1 at a later device function. With
this, my fxp card works in any slot on my 7043-140.
locators for uhub because a hub can't have sub-devices.
This might be sanity-checked eventually.
Same for ubt now after the change to device attachment.
(/dev/nvram) and implement all the associated ioctls fully. Tested with
a hacked up copy of eeprom(8). Right now it can only be used to see the
nvram GEV contents, not actually edit them. Will do that later some day.
- finish implementing splraiseipl (and makeiplcookie).
http://mail-index.NetBSD.org/tech-kern/2006/07/01/0000.html
- complete workqueue(9) and fix its ipl problem, which is reported
to cause audio skipping.
- fix netbt (at least compilation problems) for some ports.
- fix PR/33218.
long standing issue with interrupts onthe PowerStack E1. Remove the
quirk entry for the PowerStack E1 and revert to using the IVR.
Also, add a \n to the attachment of the mk clock printf that was missing.
by the initial PCI_NETBSD_CONFIGURE, nor should it be manually wired by
the indirect stuff. This is all taken care of by the prep_init() function.
7025-F40 support supports the primary pci bus only. The second PCI bus
on the machine is still unsupported, but the machine runs and can access
everything except for the two slots the other bus covers.
The bug displayed itself by locking up console output on a 7043-140
following a powercycle. This may also have been the cause of bogus
interrupts on motorola class machines.
for quite some time. Add it to all systems that have pcmcia SCSI.
Pointed out by Björn Johannesson <rherdware@yahoo.com> in private mail,
OK'd by matt@
try to locate it using the residual data. On some machines this doesn't
work, so we fall back to hardcoding it. This should fix isa interrupts
on the 7025-F40, which has the IVR at a different location.
out how to properly reboot these machines, instead we make use of the
auto-poweron-alarm time, and power cycle the machine to simulate a
reboot. This is a hack, but until I get documentation on these machines,
at least you will be able to reboot them.
for a magical PCI device location (that is sometimes wrong), we scan the
residual data for an MPIC, and if we find one, wire it up from there.
This will hopefully allow interrupts to work on the MPC750 and the
7025-F40. I suspect however the interrupt vector address on the 7025 will
still need some work.
1) Fix support for the powerstack E1. This machine needs to use the 8259
directly, and cannot use the prep interrupt vector register. Place a
quirk entry in the table for the machine.
2) Add a new com0_vreset boot image. The vreset code only works on a few
machines, and breaks others like the 7025-F40. Its only limitedly useful
when used with the com0, so just make it an optional image the user can
install by hand if they want.
3) Bump the bootloader to 1.8 with the above change.
2) Modify pnpbus attachment code to record the chipid of the device if it
has one.
3) Change the clock probes to use the chipid, rather than relying on
potentially untrustworthy subtype and interface.
4) Add decoding of memory ranges to the RESIDUAL_DUMP code.
5) Add a we@pnpbus device to allow netbooting and root device detection
from an IBM we ethernet. (it will only work if your firmware detects it)
6) Because I moved the pnpbus probe to occur prior to pci and isa, it
screwed up the root device detection and firmware path building code.
Completely rewrite the fw-path detection code to deal with this.
and INSTALL_SMALL. Since floppy images have to fit on one floppy, we
have to really cut down on the drivers on that image, but people
netbooting don't have that problem, so they deserve a more featured
kernel to install with. (namely one with raid support and more ethernet
drivers)
ntp on my 7248, however, my 7043-140 is still a bit flaky. I suspect the
only way to fix the 7043 is going to be writing a timecounter driver for
the 8254 present on these machines. Either way, this makes some of the
machines better, and the other machines are still about the same as they
were before, so it's a net gain for the port.
residual with the pnpbus probes, than it is to do it with raw isa probes,
so I've replaced the isa mkclock and mcclock code with a pnpbus attachment.
While writing the mkclock code, I realized that on motorola prep machines
the mkclock uses the same port range as the nvram part. (it's actually
the same chip/part). This was causing the nvram not to work on those
machines. Now the nvram code will recognize this, and wire up the
mkclock as well. The mkclock probe is just a stub probe used to
pre-detect the fact that this is one of those machines.
1) Rather than build an array of the pci->intr mappings, build a proplib
tree.
2) add a SIMPLEQ of device properties to the pci_chipset_tag to hold the
proplib trees of each bus.
3) Move the interrupt routing code from pci_conf_interrupt to pci_map_intr()
4) Deal properly with non-native PCI bridge chips that are not recognized
by the prep firmware, and therefore are not noted in the residual data.
The major win of this restructure is #4. Hopefully I haven't broken
anything. Tested on a 7248-100, 7043-140 and 7024-E20.
proper pcipnp structure file to read them rather than magical array offsets.
Add more complete decoding of the VPD, including extendedvpd, and TLB. While
I was here, fix a bug where we never printed the L1 cache data properly.
models. This code was made possible by assistance from Cory Bajus.
Add code that asks the PCI bridge what it's config base address is, and
use that when wiring up an indirect bridge type.
Move prep's user segment register to 10, because the 7025's PCI config
address is at 11, where the user segment used to be.
Add a battable entry for 0xbf800000 for machines with RS6K bridges.
There is still probably work left to be done before a 7025-F40 is fully
supported.
1) The E1 seems to have the int. siop wired to irq 14-level and the internal
wdc wired to irq 14-edge. special case and fail the wdc probe on E1's.
2) If we fail to map the NVRAM registers, return, rather than trying to talk
to them and panic'ing the box.
3) revert my previous "fix" to pnpbus to make irq's default to level. It
was wrong, and didn't even fix the powerstack.
With this, we have limited PowerStack E1 support. The machine cannot
talk to it's IDE controller, and cannot detect it's boot device
automatically, but it does come up and run. Tested with NFS root.
1) Rewrite vreset.c completely. Saves about 1k, and now it sets the
screen up much better. Previously the resulting video mode looked bad on
my CRT, and my LCD wouldn't even recognize it as a valid mode. Now it
looks reasonably good and my LCD can talk to it.
2) Call vga_reset unconditionally in the setup process to fix the onboard
VGA. If you have a VGA and use the serial console, this will allow you
to still use the VGA device for getty/wscons. Tested on a 7248 with VGA
and a 7043 without VGA.
3) Change filesystem.c to use nullfs rather than ufs and cd9660. We
don't actually need an fs at all, but it would take a major rewrite to
get rid of them completely. This saves about 6KB on the bootloader.
4) Bump version to 1.7.
to the screen on which they are being called. The driver cannot guess
this by itself but it is needed to implement, at least, the getwschar and
putwschar functions in the correct place. There are no functional changes
yet.
Tested on i386 (vga, vga_raster, machfb, vesafb), macppc and sparc64.
Suggested and reviewed by macallan@.
set the makefile to build the bootloader without the DB_MONITOR stuff by
default, as it is primarily for debugging. Reduces bootloader size by
about 8k.
ones and those for specific machines of developers. PR 32304.
OK'ed by rpaulo.
N.B. stf is a cloning device, so it still must be enabled by
"ifconfig stf0 create".
It now handles three different cases:
1) if you have an old disk.. it reads the partition table, and writes the new
image with that.
2) if you are doing a new install, you can build an image w/o the table that
can be dd'd to any PReP boot partition.
3) it works the old way for making netboot or floppy images
Additionally, it now includes libz, and automatically compresses the
kernel image, rather than requiring an external step.
1) Add the NVRAM device. This device allows us to speak with the nvram on
prep-based machines and read/write to it. Also add a simple IOCTL
interface for speaking with the nvram from userland. This hasn't been
tested yet, but eventually I plan to support it with the sparc eeprom
command.
2) Change the root device detection to use the nvram device to attempt to
guess the boot device. Most machines should now correctly guess thier
boot device, though I expect a few devices to still not work quite right yet.
3) change the default IRQ to level rather than edge in the pnpbus if the
flags are invalid or empty. (based on output from a PowerStack E1)
4) correctly handle older machines in pnpbus that have FixedIOPorts
rather than variable ones.
Still have much to do.
1) create new pnpbus psuedo bus. This is a bus layer that reads the PNP
tree from the residual data and allows attachment of devices with the
information given therein. Based loosely on i386/pnpbios.
2) Delete obio bus, as with the pnp bus we no longer need it.
3) Create a number of functions that gather the information needed to set
up the machine from the residual data, rather than hardcoding it in.
4) Create a quirk table for machines that are bizzare enough that the
residual information is not sufficient. (such as the 6015)
5) Using the data gathering routines and the quirk table, delete struct
platform completely from the architecture. Prep is now almost completely
dynamic in figuring out the machine it is running on and setting things
up properly.
6) Add a wdc_pnpbus driver which attaches the wdc controller found on
some 7248's and the 6015. This replaces the now-defunct wdc_obio.
7) delete all the mot_* and ibm_* files, and replace them with a single
ibm_machdep.c which only contains the quirk functions for the 6015 and
the 6050.
8) Modify GENERIC to work with all this stuff.
Changes:
Added a pci-EISA bridge device and associated glue to the prep port.
Adapted from i386.
Change pci_intr_fixup_*() to pass the pin argument on all platforms.
Make PCI_NETBSD_CONFIGURE leave the MPIC alone. We map it separately.
7024 kernel was also succ. tested on an IBM 7043, so no regression there.
a tab to two spaces. This allows more space for some of the large vendor
items we wish to print.
2) Add decoding of a number of large vendor items of particular interest
to anyone wishing to port NetBSD to a prep machine.
3) Add a small convenience function which will be used by other pnp functions
in the future.
- initialize it properly in the bus front-ends (all 2, exept in wdc_pcmcia.c
for the "Sandisk CompactFlash Card" where it's set to 1)
- remplace hardcoded '2' by ata_channel->ch_ndrive in MI IDE drivers.
From Christos Zoulas in kern/32501.
was developed as part of Google's Summer of Code 2005 program. This
change adds the kernel code, the mount_tmpfs utility, a regression test
suite and does all other related changes to integrate these.
The file-system is still *experimental*. Therefore, it is disabled by
default in all kernels. However, as typically done, a commented-out
entry is added in them to ease its setup.
Note that I haven't commited the required mountd(8) changes to be able
to export tmpfs file-systems because NFS support is still very unstable
and because, before enabling it, I'd like to do some other changes.
OK'ed by my project mentor, William Studenmund (wrstuden@).
