We changed several interrupt numbers in Qemu to better reflect real
world hardware. Also, since we're now using proper interrupt maps, we
need to make sure we specify interrupts in the way the respective map
requires it.
Signed-off-by: Alexander Graf <agraf@suse.de>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@680 f158a5a8-5612-0410-a976-696ce0be7e32
There is a draft spec that defines how interrupts are supposed to be mapped
from one bus to another, down to the PIC and CPU interrupt line.
So far we don't implement that spec. But PPC64 Linux requires us to for PCI
devices. So let's create a map here.
Draft: http://playground.sun.com/1275/practice/imap/imap0_9d.pdf
Signed-off-by: Alexander Graf <agraf@suse.de>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@679 f158a5a8-5612-0410-a976-696ce0be7e32
When running on Qemu (TCG) time goes by at a constant frequency which
accidently happens to match with Linux's fallback value.
As soon as we go to KVM, the time base suddenly starts going as fast as
the host CPU's time base. So we need to make sure Linux knows about
it, so it can do its timekeeping properly.
Signed-off-by: Alexander Graf <agraf@suse.de>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@678 f158a5a8-5612-0410-a976-696ce0be7e32
A real 970FX has more icache, so we should reflect that.
Signed-off-by: Alexander Graf <agraf@suse.de>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@677 f158a5a8-5612-0410-a976-696ce0be7e32
This change fixes node unit address matching on arch
where cell is not 4 byte integer (tested with sparc64.)
Since we encode "reg" property chunks with encode-int
we need to use the same encoding while preparing unit
phys addr for comparison at path resolution time.
- (exact-match): calculate phys addr length using 4 byte
member encoding with /l* not cells
- find-child: encode phys addr components similar to
encode-int with 4 byte stores
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@674 f158a5a8-5612-0410-a976-696ce0be7e32
This requires an updated QEMU.
Also use little endian access ASIs.
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@671 f158a5a8-5612-0410-a976-696ce0be7e32
Reading the interrupt status register clears also status register, therefore
the interrupt status must be read only after reading the status.
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@666 f158a5a8-5612-0410-a976-696ce0be7e32
Build broke with:
GEN openbios.dict
undefined word.
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@660 f158a5a8-5612-0410-a976-696ce0be7e32
the private wordlist by default. This is required for executing Milax Fcode which defines package words which need to be externally
visible.
As a consequence, it is now possible to remove lots of "external" words used building the device tree since this is now the
default.
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@655 f158a5a8-5612-0410-a976-696ce0be7e32
- claim both physical and virtual address ranges
while walking boot memory mappings
Signed-off-by: Igor V. Kovalenko <igor.v.kovalenko@gmail.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@651 f158a5a8-5612-0410-a976-696ce0be7e32
Also pass physical addresses instead of bus addresses to VGA init.
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@648 f158a5a8-5612-0410-a976-696ce0be7e32
time.
I finally managed to track things down and it seems like we're missing
some properties in several devices.
While at it, I also bumped the ESCC IRQs to 0x24 / 0x25, as that's what
qemu now maps them on.
This patch (plus the qemu set) gets PPC64 Linux booting with
console=ttyPZ0.
Signed-off-by: Alexander Graf <agraf@suse.de>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@646 f158a5a8-5612-0410-a976-696ce0be7e32
only onedoing that, as that (really easy) case got broken by the cross
compilation overhaul.
Let's allow host = target gcc configurations again.
Signed-off-by: Alexander Graf <agraf@suse.de>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@645 f158a5a8-5612-0410-a976-696ce0be7e32
transactions even if they do not have any
i/o or memory bars. E.g. PCI spec, page 297,
gives an example of such a device:
Programming interface 0000 0000b
VGA-compatible controller. Memory
addresses 0A 0000h through 0B
FFFFh. I/O addresses 3B0h to 3BBh
and 3C0h to 3DFh and all aliases of
these addresses.
While bios could check for these devices and special-case them, it is
easier to fix this by enabling i/o and memory space unconditionally:
devices that do not support it will just ignore this setting.
Signed-off-by: Michael S. Tsirkin <mst@redhat.com>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@643 f158a5a8-5612-0410-a976-696ce0be7e32
later). This allows us to proceed further with SPARC64 boot.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@642 f158a5a8-5612-0410-a976-696ce0be7e32
this new information to correctly resolve destination (b<mark) references. See the email archives for further discussion on why
this is required.
Hopefully this should finally resolve the Fcode backward branch problem; at least all previous incorrect resolutions now appear
correct and the Fcode evaluator no longer crashes or gets stuck in a loop while attempting to boot Milax.
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@641 f158a5a8-5612-0410-a976-696ce0be7e32
According to the specification, the destination for a backwards Fcode branch must be resolved from the bottom rather than the
top of the cstack. The existing version of the code was simply doing a swap, and so nesting any branches within a backward branch
would fail since the wrong destination would be resolved from the stack.
This patch adds a new cstack-startdepth variable to keep track of the cstack base location within an execution context
(setup-tmp-comp and execute-tmp-comp) and alters the backward branches to make use of it.
With this patch in place, Milax under Qemu doesn't crash anymore but sits in an infinite loop reading sectors from the CDROM.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@638 f158a5a8-5612-0410-a976-696ce0be7e32
are not in forth stack order but in reversed (heh!) order. Our implementation
confused this.
This second patch is fairly straightforward; probably the only unexpected part is
the need to rename the existing /openprom/client-services "claim" and
"release" words to "cif-claim" and "cif-release" respectively. This is
because we need to use the "claim" and "release" words in
forth/system/ciface.fs to reverse the argument order before calling the real
underlying words.
Signed-off-by: Mark Cave-Ayland <mark.cave-ayland@siriusit.co.uk>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@637 f158a5a8-5612-0410-a976-696ce0be7e32
This is the generic command, it needs support from the filesystem
packages.
Signed-off-by: Laurent Vivier <Laurent@vivier.eu>
git-svn-id: svn://coreboot.org/openbios/trunk/openbios-devel@630 f158a5a8-5612-0410-a976-696ce0be7e32
