To support full 64-bit addresses, there is a new field `ext_lfb_base`
since Linux 4.1. It is unclear, however, how a loader is supposed to
know if the kernel is compatible with this. Filling these previously
reserved bits doesn't hurt, but an old kernel would probably ignore
them and not know that it's handling a clipped, invalid address. So
we play safe, and only allow 64-bit addresses for kernels after the
2.15 version bump of the boot protocol.
Change-Id: Ib20184cf207f092062a91ac3e6aa819b956efd33
Signed-off-by: Nico Huber <nico.h@gmx.de>
Co-authored-by: Bill XIE <persmule@hardenedlinux.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/76479
Reviewed-by: Bill XIE <persmule@hardenedlinux.org>
Reviewed-by: Tim Wawrzynczak <inforichland@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Translate the coreboot framebuffer info from coreboot tables to
the Linux zero page.
Tested in QEMU/Q35 with a kernel w/ efifb enabled.
Change-Id: I2447b2366df8dd8ffe741c943de544d8b4d02dff
Signed-off-by: Nico Huber <nico.h@gmx.de>
Co-authored-by: Bill XIE <persmule@hardenedlinux.org>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/76431
Reviewed-by: Paul Menzel <paulepanter@mailbox.org>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <inforichland@gmail.com>
Reviewed-by: Samuel Holland <samuel@sholland.org>
Reviewed-by: Bill XIE <persmule@hardenedlinux.org>
With LinuxBoot Linux relied on the legacy method of fetching the RSDP
pointer to get ACPI. This uses a more modern approach available since
2018 on the Linux kernel, which involves filling in the zero page.
This method takes precedence over any other method of fetching the
RSDP in Linux (UEFI, Kexec, Legacy/BIOS). Some UEFI zealots are
threatening that legacy code will be removed from Linux so it's best
to already adapt to that possibility.
Tested on Qemu:
- With qemu the RSDP is always in the EBDA, so checking if Linux uses
the provided pointer is better done with a forced bad entry
- With a fake bad pointer Linux correctly does not find RDSP
Change-Id: I688b94608b03b0177c42d2834c7e3beb802ae686
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/62574
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
The e820 type don't fully match the LB_TAG_MEMORY types, so change all
unknown types to e820 to '2', reserved memory.
TESTED with Linuxboot: e820 now shows the CBMEM region as reserved.
Change-Id: Ie0e41c66e002919e41590327afe0f543e0037369
Signed-off-by: Arthur Heymans <arthur@aheymans.xyz>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/55074
Reviewed-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@gmail.com>
Reviewed-by: Rocky Phagura
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
The linux trampoline was modifying the existing GDT
to add the 0x10 and 0x18 descriptors for Linux.
This will not work when the existing GDT is in ROM.
Change the code to set up a new GDT in what we know to be
RAM.
Tested by booting a linux payload. The main reason this works
is that Linux almost immediately loads its own GDT and then
segment registers. This GDT is a very temporary bridge.
Note that none of this change used to be necessary; the coreboot
GDT was originally compatible with Linux (ca 2000);
then Linux changed.
Change-Id: I13990052fbfd6a500adab8a2db8f7aead1d24fa6
Signed-off-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-on: https://review.coreboot.org/27529
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
It's not needed, so we can remove some extra file mangling, too.
Change-Id: I80d707708e70c07a29653258b4cb6e9cd88d3de3
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/12508
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Add the code necessary to create the linux trampoline blob.
Don't enforce this for the in-coreboot build or use objcopy
to produce linux_trampoline.o as it is a bit trickier to get
all the details right than I had hoped:
- you have to know the elf architecture of the host machine
- you might have to have more tools (xxd, perl, etc) installed
Change-Id: I9b7877c58d90f9fb21d16e0061a31e19fffa2470
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/12505
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
It encourages users from writing to the FSF without giving an address.
Linux also prefers to drop that and their checkpatch.pl (that we
imported) looks out for that.
This is the result of util/scripts/no-fsf-addresses.sh with no further
editing.
Change-Id: Ie96faea295fe001911d77dbc51e9a6789558fbd6
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11888
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
As per discussion with lawyers[tm], it's not a good idea to
shorten the license header too much - not for legal reasons
but because there are tools that look for them, and giving
them a standard pattern simplifies things.
However, we got confirmation that we don't have to update
every file ever added to coreboot whenever the FSF gets a
new lease, but can drop the address instead.
util/kconfig is excluded because that's imported code that
we may want to synchronize every now and then.
$ find * -type f -exec sed -i "s:Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, *MA[, ]*02110-1301[, ]*USA:Foundation, Inc.:" {} +
$ find * -type f -exec sed -i "s:Foundation, Inc., 51 Franklin Street, Suite 500, Boston, MA 02110-1335, USA:Foundation, Inc.:" {} +
$ find * -type f -exec sed -i "s:Foundation, Inc., 59 Temple Place[-, ]*Suite 330, Boston, MA *02111-1307[, ]*USA:Foundation, Inc.:" {} +
$ find * -type f -exec sed -i "s:Foundation, Inc., 675 Mass Ave, Cambridge, MA 02139, USA.:Foundation, Inc.:" {} +
$ find * -type f
-a \! -name \*.patch \
-a \! -name \*_shipped \
-a \! -name LICENSE_GPL \
-a \! -name LGPL.txt \
-a \! -name COPYING \
-a \! -name DISCLAIMER \
-exec sed -i "/Foundation, Inc./ N;s:Foundation, Inc.* USA\.* *:Foundation, Inc. :;s:Foundation, Inc. $:Foundation, Inc.:" {} +
Change-Id: Icc968a5a5f3a5df8d32b940f9cdb35350654bef9
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/9233
Tested-by: build bot (Jenkins)
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
The Linux trampoline code does not set up the segment descriptors for
__BOOT_CS and __BOOT_DS as described in the Linux kernel
documentation:
... a GDT must be loaded with the descriptors for selectors
__BOOT_CS(0x10) and __BOOT_DS(0x18); both descriptors must be 4G
flat segment; __BOOT_CS must have execute/read permission, and
__BOOT_DS must have read/write permission;
This is not a problem when launching a Linux payload from coreboot, as
coreboot configures the segment descriptors at selectors 0x10 and
0x18. Coreboot configures these selectors in the ramstage to match
what the Linux kernel expects (see
coreboot/src/arch/x86/lib/c_start.S).
When the cbfs payload is launched in other environments, SeaBIOS for
example, the segment descriptors are configured differently and the
cbfs Linux payload does not work.
If the cbfs Linux payload is to be used in multiple environments
should the trampoline needs to take care of the descriptors that Linux
requires.
This patch updates the Linux trampoline code to configure the 4G flat
descriptors that Linux expects. The configuration is borrowed from
the descriptor configs in coreboot/src/arch/x86/lib/c_start.S for
selectors 0x10 and 0x18.
The linux_trampoline code is slightly refractored by defining the
trampoline entry address, 0x40000, as TRAMPOLINE_ENTRY_LOC. This
definition is moved into a separate header file, linux_trampoline.h.
This header file is now included by both the trampoline assembly
language code and the trampoline loader C code.
The trampoline assembly language code can now use TRAMPOLINE_ENTRY_LOC
as scratch space for the sgdt CPU instruction.
Testing Done:
Verified the Linux payload is booted correctly in the following
environments:
1. Coreboot -> Linux Payload
2. Coreboot -> SeaBIOS -> Linux Payload: (previously did not work)
Change-Id: I888f74ff43073a6b7318f6713a8d4ecb804c0162
Signed-off-by: Curt Brune <curt@cumulusnetworks.com>
Reviewed-on: http://review.coreboot.org/6796
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Now that we can set CC to an arbitrary compiler, fix issues that clang
finds. Luckily, there were only two trivial errors.
Change-Id: I0fd1f0f263a8ab7004f39cd36ed42d1a1cba5c04
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: http://review.coreboot.org/5081
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
In the great tradition of LinuxBIOS this allows adding
a kernel as payload. add-payload is extended to also
allow adding an initial ramdisk (-I filename) and a
command line (-C console=ttyS0).
Change-Id: Iaca499a98b0adf0134e78d6bf020b6531a626aaa
Signed-off-by: Patrick Georgi <patrick.georgi@secunet.com>
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/3302
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Nico Huber