The touched workaround for Sandy Bridge reserves two memory regions that
could cause graphics corruption if mapped by the integrated graphics
device. To the best of our knowledge, the workaround is not needed for
Ivy Bridge revisions.
Tested on kontron/ktqm77 (Ivy Bridge): Booted Linux and checked the
memory regions are not reserved. Couldn't test on Sandy Bridge, due to
lack of hardware.
Change-Id: I4273d1d804b490cf93c23426782eb1ffaf29f7d4
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12326
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Tested-by: build bot (Jenkins)
There has been a concerted effort to clean up coreboot's microcode
handling that has included a move away from coreboot-specific
microcode file collections. As a result, the ability to specify
a single microcode file to be added to the image is of less utility
than before.
NOTE: This patch remove the built-in external microcode feature,
however the user can still specify no microcode during build and
manually add the correct microcode file(s) to the CBFS image after
the build is complete.
Change-Id: Ifea94c21e531a74953f5a0e2f489378c20ef3b5c
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/11903
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
The last_boot NVRAM option was deprecated and removed in
commit 3bfd7cc6. Remove the last_boot option from all
affected mainboards to eliminate user confusion.
Change-Id: I7e201b9cf21dfe5dda156785bad078524098626d
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12316
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins)
Most AMD hardware requires at minimum two warm resets
when booting from S5 (power off). This is uncomfortably
close to the maximum bootblock execution count, and has resulted
in unstable normal/fallback operation on some machines.
Increase the default max bootblock execution count before fallback
to 6. This translates to roughly 2 - 3 failed boots before fallback
mode will engage, with an absolute worst case of pushing the reset
button 5 times to engage fallback mode in the absence of a dedicated
recovery jumper.
Change-Id: I1911f1b77f168835b516e6a915d5b6949f47219a
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12317
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
On mingw, the function glob has some default options
which are not compliant with man page.
If gl_offs is not set as 0, there may be some slots which
is reserved.
If gl_pathc or gl_pathv is not set as 0, the result might
be appended to the list instead of being added as new ones.
Change-Id: I03110c4cdda70578828d6499262a085a81d26313
Signed-off-by: Zheng Bao <zheng.bao@amd.com>
Signed-off-by: Zheng Bao <fishbaozi@gmail.com>
Reviewed-on: http://review.coreboot.org/11711
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Tested-by: build bot (Jenkins)
The buildgcc makefile was using the variable 'BUILDJOBS' to pass the
number of cores to use for the build into buildgcc. This is changed
to 'CPUS' to match the variable name for the what-jenkins-does target.
Change-Id: I373c4988e9f096ca2e142afdd5e94d7d806891e3
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: http://review.coreboot.org/12299
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Add a DMAR table to advertise IOMMU and IRQ remapping capabilities to
the OS.
Tested with kontron/ktqm77. Under Linux, the table is detected and
interrupt remapping is enabled automatically.
Change-Id: Id6ee601a0a8543ed09c6bb8d308a3a3549fc34e5
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12195
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Sandy Bridge and Ivy Bridge processors have two IOMMU units. One for the
integrated graphics controller and one for all other PCI devices. Assign
resources for both IOMMUs and apply some quirks.
Tested with kontron/ktqm77 and a Muen based system that makes use of the
IOMMUs. Not tested on Sandy Bridge, but register dumps show the same
settings that are applied here.
Change-Id: I43b5e20b750e7529f448acac35de173185678fd9
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12194
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Assign unique bus/dev/fn values for the I/O APIC and each HPET. The
values are taken from an example DMAR table. They are used as source-id
for MSI requests and as completer-id for reads from the device' MMIO
space [1, 2]. The former is usefull for source-id verfication during
interrupt remapping.
[1] Intel 6 Series Chipset and Intel C200 Series Chipset
Datasheet
Document-Number: 324645
[2] Intel 7 Series / C216 Chipset Family Platform Controller Hub (PCH)
Datasheet
Document-Number: 326776
Change-Id: Ib46f8cfb7d966dd1cf2b026f671bc45ffcc43d25
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12193
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
On server boards with a recovery jumper, having the fallback path
less sensitive to power fluctuations or BMC issues makes sense.
Increase the maximum number of boot attempts before automatic
fallback to 10 on these boards.
Change-Id: Iabe0b0cbf332686db8e9380a8b65a1477173599c
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12320
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Refactor acpi_create_dmar_drhd_ds_pci() and add similar functions for
I/O-APICs and MSI capable HPETs. We violate the spec [1] here, which
talks about 16-bit source-ids spread over start_bus and path entries.
Intel actually uses bus/dev/fn identification for those devices too,
and so do we.
[1] Intel Virtualization Technology for Directed I/O
Architecture Specification
Document-Number: D51397
Change-Id: I0fce075961762610d44b5552b71e010511871fc2
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12192
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Add a parameter to acpi_create_dmar() for the flags field and define
flags given by the spec [1].
[1] Intel Virtualization Technology for Directed I/O
Architecture Specification
Document-Number: D51397
Change-Id: I03ae32f13bb0061bd3b9bef607db175d9b0bc5e1
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12191
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Duncan Laurie <dlaurie@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
All boards using this SuperIO have been removed from the tree already.
Change-Id: I52847bc2fc16b27ac0de0bc7c847221b1e5cb744
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: http://review.coreboot.org/12245
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
The cbmem utility shouldn't be using the intra coreboot
data structures for obtaining the produced data/information.
Instead use the newly added cbmem records in the coreboot
tables for pulling out the data one wants by using the
generic indexing of coreboot table entries.
BUG=chrome-os-partner:43731
BRANCH=None
TEST=Interrogated cbmem table of contents with updated code.
Change-Id: I51bca7d34baf3b3a856cd5e585c8d5e3d8af1d1c
Reviewed-on: http://review.coreboot.org/11758
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Fix a function call in the normal path using the original function
name and arguments in code that was changed in commit 3bfd7cc6
(drivers/pc80: Rework normal / fallback selector code)
This commit reworked most of the fallback / normal code,
however the normal code paths were not fully tested by Jenkins,
so this was missed.
Change-Id: Ied66334977272a13b7a7307ff4d9f34eb22040aa
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12315
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins)
Per IRC and Gerrit discussion, the normal / fallback
selector code is a rather weak spot in coreboot, and
did not function correctly for certain use cases.
Rework the selector to more clearly indicate proper
operation, and also remove dead code. Also tentatively
abandon use of RTC bit 385; a follow-up patch will
remove said bit from all affected mainboards.
The correct operation of the fallback code selector
approximates that of a power line recloser, with
a user option to attempt normal boot that can be
cleared by firmware, but never set by firmware.
Additionally, if cleared by user, the fallback
path should always be used on the next reboot.
Change-Id: I753ae9f0710c524875a85354ac2547df0c305569
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12289
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Show laptops and servers before desktop boards since that's where both
the market and coreboot are the most active these days.
Change-Id: I7de63975f3f2ff5e983b19e07558175a58870a1b
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12292
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
There's the sentiment that the Supported_Motherboards wiki page is
outdated. Point out that the list is current (and drop the table of
contents that became a distraction).
Change-Id: Ib2363fad0b7f6951b07b2ad0c85148d9bc729b55
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12291
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
The K8 PowerNow! state generator does not generate _PSS objects
for nodes other than the first CPU package. This patch backports
the PowerNow! core count fixes for Family 10h to the K8 CPUs.
Change-Id: I7b411ab75155dfb4bf51ae04301aa16fb2ae89f3
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12286
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Change-Id: I0180e0ae2aeeffcef46a97892356f1955f581efd
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: http://review.coreboot.org/12295
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
In order to not expose the cbmem data structures to userland
that are used by coreboot internally add each of the cbmem
entries to a coreboot table record. The payload ABI uses
coreboot tables so this just provides a shortcut for cbmem
entries which were manually added previously by doing the
work on behalf of all entries.
A cursor structure and associated functions are added to
the imd code for walking the entries in order to be placed
in the coreboot tables. Additionally a struct lb_cbmem_entry
is added that lists the base address, size, and id of the
cbmem entry.
BUG=chrome-os-partner:43731
BRANCH=None
TEST=Booted glados. View coreboot table entries with cbmem.
Change-Id: I125940aa1898c3e99077ead0660eff8aa905b13b
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: http://review.coreboot.org/11757
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
TEST: Booted ASUS KGPE-D16 with single Opteron 6380
* Unbuffered DDR3 DIMMs tested and working
* Suspend to RAM (S3) tested and working
Change-Id: Idffd2ce36ce183fbfa087e5ba69a9148f084b45e
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/11966
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
To store memory configuration in SPI flash currently adds
some 150 ms delay in ramstage, visible in timestamps listing
at 75:cbmem post.
Change-Id: I1160259054b58e9a8df2a105c730e0f4140be1f5
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/12215
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Disables mouse ps2 data/clock signals, not connected in hardware.
Purpose of other GPIOs is not really known, but match them
with superiotool dump taken from vendor bios.
Change-Id: I7b549fbd7dd3fa4cbd507d76882b60bc324a4bd0
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/12214
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
While the actual pins behind these devices are not exposed on the chip,
the enable registers are implemented in hardware. Allow to turn these LDNs
off, like the vendor bios for asrock/e350m1 does.
Change-Id: I4d6d5a8de12b09095138cacbad62b2dfbbe54028
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: http://review.coreboot.org/12213
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Several ramstage files were inadvertently using Family 10h-specfic
structures, causing unstable operation. Use the K8-specific
structures instead.
Change-Id: I64066dfdca83557393499b77726051e25b814381
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12290
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
abuild -t EMULATION_QEMU_UCB_RISCV,EMULATION_SPIKE_UCB_RISCV works now
Change-Id: I49d8cd86e21ede724d8daa441b728efa1f6ea1fa
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12281
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
junit reports were kept around (and appended to) in some cases, leading
to duplicate reports on jenkins.
Drop old per-mainboard reports before building said boards, and do the
same for the tools (reported thrice).
Change-Id: I74a035587bbf917dca85ba6fc74621c583efe9a2
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12280
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Specifying a directory with multiple boards (eg abuild -t google/veyron)
makes abuild run through all of them.
Change-Id: Ifb60f3a1f0c4a727dc43c48671ea90711ffe5585
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12278
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Since we now have multiple boards in a single mainboard directory (eg
google/veyron), we need some other identifier from which to create
output directories and filenames in abuild than the directory name.
Use the wildcard part of CONFIG_BOARD_* instead.
This changes the semantics of payload.sh handling: it's passed the
single new identifier instead of two arguments "vendor" and "board" that
constitute the mainboard directory's path.
Change-Id: I0dc59c6a1ad1ee51d393fa06b98944a6da342cdf
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12277
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
It only takes a single argument now, which is the directory below the
coreboot-builds directory. Preparation for future work.
The only visible change is in console output.
Change-Id: I4b0fe268ccfb69a0403fa5f8b23444c07843386f
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12276
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
It's passed the mainboard's directory name (below $TARGET) directly
in preparation of more rework in that area.
Change-Id: I3a82b8673fdea07bc5c957f76f4685c34a805334
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12275
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Its hardcoded HTTP endpoint is gone since 2007.
Change-Id: Ib76814d31b571456d950d45f45912036b6fa82d1
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12274
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
If you already have a configuration, there's no need to run it through
abuild.
Change-Id: I4dde9a7b96bb0c08ec5c91426a4dd3aa15e74edf
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/12273
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The acpi_get_sleep_type function in SB700 ramstage is only needed
for boards / CPUs that require late CBMEM initialization.
Providing this function in early CBMEM-compatible boards breaks
building of the ACPI S3 code due to multiple definitions of
acpi_get_sleep_type.
Change-Id: Ieebc2640a586812e3e2bfd410987205d64147314
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12267
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Martin Roth <martinroth@google.com>
One of the interrupts in intel_vga_int15_handler lacks
positive return status. Write correct status to avoid
error messages in log.
TEST=With this change `int15 call returned error` is not shown anymore
on a custom board with Intel Atom CPU, i945GME northbridge and
i82801gx southbridge.
Change-Id: I740b2df9bd6a7d261d89bef74b924edbb64354aa
Signed-off-by: Konstantin Aladyshev <aladyshev@nicevt.ru>
Reviewed-on: http://review.coreboot.org/12255
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
checkpatch.pl that we inherited from Linux checks for its absence, so it
may be easiest to follow their style of not caring for the FSF's address
anymore.
TEST=visual check that `git diff` and `git diff |grep "^[+-]" | \
grep -v "^--- " |grep -v "^+++ " |sort | uniq -c |sort -n` look
reasonable (matching number of removed and added comment terminators */,
etc.).
Also, `git grep -A3 "You should have received a copy"` only
returns license texts, imported files, patches and help strings in
applications as remaining copies of that paragraph
Change-Id: I7c43860b6fd7ec526983c24b608994539128cfb9
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: http://review.coreboot.org/11887
Tested-by: build bot (Jenkins)
Reviewed-by: Ronald G. Minnich <rminnich@gmail.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>
The existing KFSN4-DRE support hung during ramstage while initializing
AP #3 if a second CPU package was installed. After analyzing the
Sun Ultra 40 M2 support code it became apparent that the K8 code
cannot function correctly if sequential RAM training is disabled,
and that there were a few other missing calls. This patch adds
the missing calls, adjust the CAR space to an appropriate level, and
explicitly defines the link numbers and connections in devicetree.cb
TEST: Booted ASUS KFSN4-DRE with 2x Opteron 8222 installed.
Change-Id: I96178b7367b0c13de5c9d5d90d032fb0c53639c2
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12285
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
Disable the parallel CPU initialization for model_206ax, that is Sandy
Bridge and Ivy Bridge processors. We never did it the way that Intel
recommends and it became unreliable with the introduction of SMM_MODULES
in commit a3e41c0 Migrate 206ax to SMM_MODULES.
Tested by booting kontron/ktqm77 2.6k times into Linux user space. No
issues so far.
Change-Id: Idffc352341419f22a36bf772534a5e11e711edf1
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: http://review.coreboot.org/12266
Tested-by: build bot (Jenkins)
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Nico Huber