Uppercase variable names can be confused with register definitions. Use
lowercase names instead, conforming to the coding style guidelines.
Tested with BUILD_TIMELESS=1, Packard Bell MS2290 remains identical.
Change-Id: I61a28bf964ea8c2c662539825ae9f2c88348bdba
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43730
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This is the only instance of `BETTER_MEMORY_MAP` in the tree.
Change-Id: I118e5b5a0f10da56e2335828477caed81c5bf855
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43729
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
This register does not seem to exist on Ironlake.
Change-Id: I3fba6a3fd443f2c9eab874e1d1b8f081f58b1536
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43728
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Looks like some registers are defined twice. Also, group some QPI
registers together. They were scattered around and mixed with the host
bridge registers, probably because other northbridges have such
registers in the host bridge's PCI config space. But not Ironlake.
Tested with BUILD_TIMELESS=1, Packard Bell MS2290 remains identical.
Change-Id: I6e60f7fcb1467f302618eeab1b0d995920a98569
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43726
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
Sort them by ascending offsets.
Tested with BUILD_TIMELESS=1, Foxconn D41S does not change.
Change-Id: I521aa3e49b17a9fb6b279ae758801356e510d054
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43725
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
From a log of a machine using Crystal Well CPU [1], Crystal Well CPUs
use some new PCI IDs. Without this patch, the Crystal Well northbridge
cannot be initialized in ramstage, thus the machine cannot boot. Some
PCI IDs of Crystal Well related devices can be found in the PCI ID
database [2].
Tested with i5-4570R (with LGA1150 mod) on ASRock H81M-HDS. The board
boots to SeaBIOS with boot screen displayed on HDMI output, and then
boots Arch Linux on a USB disk.
[1] https://mail.coreboot.org/hyperkitty/list/coreboot@coreboot.org/thread/DNHLQTNTRQT43T67DG7L2HVI5CV74ZCM/
[2] https://pci-ids.ucw.cz/read/PC/8086
Change-Id: Icfe55323fd06187148c788ebfa7b679b6944e4f3
Signed-off-by: Iru Cai <mytbk920423@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/41658
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Use the name of the assembly instruction it uses, mfence.
Change-Id: I98d7926434694a41fb6415bed4276741fa7996af
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43822
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tim Wawrzynczak <twawrzynczak@chromium.org>
Fill in the maximum DRAM capacity and slot count read from CAPID0_A
registers on Sandy Bridge and Haswell.
While the register isn't part of the Core Series datasheet, it can be
found in the corresponding "Intel Open Source Graphics Programmer's
Reference" datasheets.
Note that the values for DDRSZ (maximum allowed memory size per channel)
need to be halved when only one DIMM per channel is supported. On mobile
platforms, all but quad-core processors are subject to this restriction.
Tested on Lenovo X230:
On Linux, verify that `dmidecode -t 16` reports the actual maximum
capacity (16 GiB) instead of the currently-installed capacity (4 GiB) or
the max capacity assuming two DIMMs per channel is possible (32 GiB).
Change-Id: I6e2346de1ffe52e8685276acbdbf25755f4cc162
Signed-off-by: Patrick Rudolph <patrick.rudolph@9elements.com>
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43971
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Christian Walter <christian.walter@9elements.com>
On Haswell platforms, the processor and the PCH are two separate dies,
and communicate through a high-speed bus. This is DMI (Direct Media
Interface) on traditional two-package platforms, but single-package
Haswell LP variants use OPI (On-Package Interconnect) instead.
Since OPI is not routed through the mainboard, most link parameters are
static and cannot be changed. OPI self-initializes on boot, anyway.
However, DMI needs to be initialized in firmware. On Haswell, the MRC
initializes the physical DMI link, but things like topology and power
management need to be configured as well. And we don't do that properly.
We enable ASPM on the PCH side of the DMI link, but not on the SA side.
Both sides need to use the same settings, so enable DMI ASPM on the SA.
Clearing the error status bits needs to be done on all Haswell variants.
Tested on Asrock B85M Pro4, still boots.
Change-Id: Ie97ff56eec9f928cfd2d5d43a287f3e0d2fbf3cf
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43743
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
The Kconfig lint tool checks for cases of the code using BOOL type
Kconfig options directly instead of with CONFIG() and will print out
warnings about it. It gets confused by these references in comments
and strings. To fix it so that it can find the real issues, just
update these as we would with real issues.
Signed-off-by: Martin Roth <martin@coreboot.org>
Change-Id: I5c37f0ee103721c97483d07a368c0b813e3f25c0
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43824
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
This to silent a bug found using gcc-10.
src/northbridge/intel/ironlake/raminit.c: In function 'setup_heci_uma':
src/northbridge/intel/ironlake/raminit.c:1805:11: error: 'reply.command' may be used uninitialized in this function [-Werror=maybe-uninitialized]
1805 | if (reply.command != (MKHI_SET_UMA | (1 << 7)))
| ~~~~~^~~~~~~~
cc1: all warnings being treated as errors
Change-Id: I0d13de549b6d428ac3675ee3f91eb5e42aeb25e8
Signed-off-by: Elyes HAOUAS <ehaouas@noos.fr>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/42461
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Add missing registers and sort them by ascending offsets.
Tested with BUILD_TIMELESS=1, Asrock B85M Pro4 does not change.
Change-Id: I98f836668144032d920b56afff878acc0a58ed82
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43691
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
It was only used in one function, but its value was never read. Drop it.
Change-Id: Ib511352d51d4452d666640d0f52810b06c8d61ce
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43702
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
There's no need to set up the southbridge in the northbridge code.
Change-Id: I0f80c92aca885812c27a8803c2745844d8dfb939
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43689
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Make it default to 0x400, which is what the touched southbridges use.
Change-Id: I95cb1730d5bf6f596ed1ca8e7dba40b6a9e882fe
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43037
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Nico Huber <nico.h@gmx.de>
There's no generic way to tell whether a mainboard has an EC or not.
Making Kconfig symbols for these options seems overkill, too. So, just
put them on the devicetree. Also, drop unnecessary assignments when the
board's current value is zero, as the struct defaults to zero already.
Change-Id: If2ebac5fcab278c97dfaf8adc9d1e125888acafe
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43129
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
If the Intel in-PCH GbE MAC is enabled in the devicetree, then tell MRC
to enable it as well. No one can ever forget to set this option anymore!
Change-Id: I946af36d16c94bb1a0f146604d0329fe6d6ce7e2
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43128
Reviewed-by: Nico Huber <nico.h@gmx.de>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
And use it instead of directly writing to the MRC struct.
Change-Id: I7f04db29a08512c1a8b2b2300dba71cb3b84a5c5
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43127
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Check the PCH's LPC device ID to know the system type instead of relying
on hardcoded numbers. The `get_pch_platform_type` function is MRC-safe.
Change-Id: Icfe7c2dccb7c7a178892ad3a2e34ca93b33b2bb9
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43124
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This Kconfig symbol allows doubling the memory's refresh rate, assuming
that the MRC actually cares about it. It is disabled by default except
on the mainboards which explicitly enabled this setting in `pei_data`.
Change-Id: I6318dad0350d1c506c67f9d117d0ae8dad871281
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43122
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
All mainboards have a non-zero SPD address to implemented DIMM slots.
Knowing this, it is possible to compute the MRC slot population masks
automatically instead of hardcoding the values on each mainboard.
Change-Id: Ia8f369dd1228d53d64471e48700e870e01e77837
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43119
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Reviewed-by: Michael Niewöhner
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
These settings are the same on all boards. Since the other boards
currently overwrite the struct contents, it doesn't make a difference.
To ease review, the same settings will be dropped from other boards in
separate commits, one board at a time.
Change-Id: I500b7a1d7d97c6976e0c7c10ca491d3875cae22b
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43109
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
This is what sandybridge does, and if done properly allows factoring out
common settings. Said refactoring will be handled in subsequent commits.
Change-Id: I075eba1324a9e7cbd47e776b097eb940102ef4fe
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43108
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
It only contains a pointer to another struct. Flatten it.
Change-Id: Iab427592c332646e032a768719fc380c5794086b
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43106
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Instead of using function pointers, we can use weak functions. So, drop
the pointer from `romstage_params`, leaving `pei_data` as the only
remaining member. This will be cleaned up in a follow-up commit.
Change-Id: I3b17d21ea7a650734119a5cab4892fcb158b589d
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43105
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
This function is called at the end of `romstage_common`. Only one board
makes use of it, the Lenovo ThinkPad T440p. To preserve behavior, call
it after `romstage_common` has done nearly everything.
Change-Id: I35742879e737be4f383a0e36aecc6682fc9df058
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43094
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Code has evolved such that there seems to be little
use for global definition of cbmem_top_chipset().
Even for AMD we had three different implementations.
Change-Id: I44805aa49eab526b940e57bd51cd1d9ae0377b4b
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43326
Reviewed-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-by: Nico Huber <nico.h@gmx.de>
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This code is not even being build-tested. Drop it before it grows moss.
Change-Id: I5e33526a02872c14e9fa37a485d2f93dea8b088f
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43230
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michael Niewöhner
This code is not even being build-tested. Drop it before it grows moss.
Change-Id: Icb98f3535f6c5f51081fc82262f6413f4b1a5733
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43261
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Michael Niewöhner
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This code is not even being build-tested. Drop it before it grows moss.
Change-Id: I41a4f73df7fdd372ec7a80a41c8216c502054c39
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43262
Reviewed-by: Michael Niewöhner
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
This code is not even being build-tested. Drop it before it grows moss.
Change-Id: I36500c1f0eb3c37d08c691d22382ceca732d1355
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43231
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Michael Niewöhner
This simplifies things and makes type checking possible.
Change-Id: Iefc9baabae286aac2f2c46853adf1f6edf01586f
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43103
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Instead of passing around a pointer to an array, just write the relevant
registers directly. Note that intel/baskingridge used spaces to indent
line continuations and had to be replaced with tabs to quell Jenkins.
Change-Id: Ifa06a2ab24da9b8c6aac6480542fa32d04f6d6fe
Signed-off-by: Angel Pons <th3fanbus@gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/43097
Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
Reviewed-by: Tristan Corrick <tristan@corrick.kiwi>
Reviewed-by: Arthur Heymans <arthur@aheymans.xyz>
