During receiver enable cycle training on Family 15h the entire range
of possible delays is searched, even though the single passing window
is often found nearly immediately. Skip the remainder of the delay
range after the passing window has been located.
Change-Id: If98217fa8e7de77366762d3c7bb01049a1dc080f
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14544
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
During DQS receiver enable cycle training on Family 15h platforms the
read data timing registers were inadvertently set to zero on every
lane training attempt.
Ensure that the read data timing registers are correctly set after
each lane is trained in receiver enable cycle training. This allows
more than one RDIMM to function on a given DCT channel.
Change-Id: I87d732f0383e9785a73b57e6f48855f3e872f1f9
Tested-On: ASUS KGPE-D16
Tested-With: 1x Opteron 6262HE
Tested-With: 4x Crucial 36KSF1G72PZ-1G6M1 (slots A2 / A1 / B2 / B1)
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14543
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The existing Family 15h receiver enable training code stored
temporary delay values in the wrong variables, leading to
the requisite averaging of delays across nibbles not being
applied. This in turn made x4 DIMMs less stable than they
should have been.
Store temporary nibble delay values in a dedicated array.
Change-Id: Ic5da898af7d689db4110211f89b886ccdbb5f78f
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14541
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Fix regression introduced by:
Ib48fe8380446846df17d37b22968f7d4fd6b9b13
Don't run channel_test on S3 resume as it overrides memory
that might be in use.
Fixes MCE events reported by the GNU/Linux kernel that
low memory has been modified.
Reset on failed s3 resume.
Change-Id: Ibadea286619c7906225f86a93aaa0b4caf26cabe
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/14439
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
DIMM training can sporadically fail due to external influences or various
errata. In these cases, restarting to retry training is a more appropriate
response than halting the system and requiring manual intervention.
Change-Id: Id49f7419f56e0640a84448cc06ecbaf62bed145e
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14529
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The wrong DIMM number was used in the initial non-target MRS
setup routines. This had no functional impact other than to
print the wrong DIMM number in the DDR3 verbose debug output.
Change-Id: I480118ed00e1786a06e641a56f0fb19cd87f92eb
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14501
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The existing RDIMM RC control word send routines were a hodgepodge
of various AGESA chunks with different ways of handling the same
task. Unify the control word chip select setup, use precise timing
routines on Family 15h, fix a couple of incorrect masks, and add
additional debugging statements.
It is believed that this patch is cosmetic and does not significantly
alter existing functionality.
Change-Id: Ie4ec7b6a7be7fce09e89f9eec146cc98b15b6160
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14500
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
When more than one DIMM is installed on a DCT, only the first DIMM
delay values are scaled to the new memory clock frequency after a
memory clock change during write leveling.
Store the previous memory clock of each DIMM during write leveling
to ensure that every DIMM has its delay values rescaled.
Change-Id: I56e816d3d3256925598219d92783246f5f4ab567
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14479
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
After substantial testing it has been determined that it is neither
required nor safe to disable the DRAM MCA during initial startup.
This (mostly) reverts commit c094d99611.
The minor debugging enhancements from that commit were left in place.
Tested-On: ASUS KGPE-D16
Config-CPU: 1x Opteron 6262HE
Config-RAM: 4x Crucial 36KSF1G72PZ-1G6M1
Config-RAM: 1x Kingston 9965516-483.A00LF
Change-Id: I58fcc296b8c45ecaedf540951c365e4ce52baaf5
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14446
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Certain RDIMMs have inherently large write levelling delays,
in some cases exceeding 1.5 MEMCLK. When these DIMMs are
utilized, the phase recovery system requires special handling
due to the resultant offset exceeding the phase recovery reporting
capabilities.
Fix an old error where delays > 1.5 MEMCLK were not being programmed
(gross delay high bit was not in set range), and restore special
delay handling for delays greater than 1.5 MEMCLK.
Also enhance debugging for x4 DIMMs around the affected code.
Tested-On: ASUS KGPE-D16
Config-CPU: 1x Opteron 6262HE
Config-RAM: 4x Crucial 36KSF1G72PZ-1G6M1
Change-Id: I0fb5454c4d5a9f308cc735597607f095fe9188db
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14441
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The BKDG requires phy fences to be re-trained after a memory clock change.
Memory training on the ASUS KGPE-D16 and KCMA-D8 somehow "mostly" worked
-- without actually following this requirement -- !
Fix the single typo that caused several weeks of delay in putting
servers with Kingston RAM (and others) into production...
Tested-On: ASUS KGPE-D16
Config-CPU: 1x Opteron 6262HE
Config-RAM: 4x Crucial 36KSF1G72PZ-1G6M1
Change-Id: I197e6728d2b0ac8c1535740599459d080b17af33
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14445
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
We never define B1_IMAGE or B2_IMAGE. These are about building
CIMx as separate binary modules, while coreboot builds these into
same romstage or ramstage module.
Change-Id: I9cfa3f0bff8332aff4b661d56d0e7b340a992992
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/14393
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-by: Kerry Sheh <shekairui@gmail.com>
Reorder drivers to fit src/drivers/[X]/[Y]/ scheme to make
them pluggable.
Also, fix up the following driver subdirectories by switching
to the src/drivers/[X]/[Y]/ scheme as these are hard requirements
for the main change:
* drivers/intel
* drivers/pc80
* drivers/dec
Change-Id: I455d3089a317181d5b99bf658df759ec728a5f6b
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/14047
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
gcc doesn't like these because they're undefined behavior, so use
zeroptr instead. For the loop that just does a number of writes (0..4),
use zeroptr + i.
Checked the disassembly (AMD_RUMBA and PCENGINES_ALIX2D) to not contain
ud2 anymore and to look reasonable where zeroptr was used.
Change-Id: I4a58220ec9a10c465909ca4ecbe5366d0a8cc0df
Signed-off-by: Patrick Georgi <pgeorgi@chromium.org>
Reviewed-on: https://review.coreboot.org/14345
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Tested-by: build bot (Jenkins)
Trivial; Use tab over space for indent. Clean up some ASCII art
while here.
Change-Id: Id2478d140a98596c5eeefdf5b047c1ca23203909
Signed-off-by: Edward O'Callaghan <eocallaghan@alterapraxis.com>
Reviewed-on: https://review.coreboot.org/8016
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Two of the MCT data structures passed as substructures to ramstage were
not packed, and additionally no alignment was specified. On at least
SP5100-based platforms, specifying packed with no alignment caused boot
failure dependent on the exact compiled binary layout (LPC hang).
Specifying the alignment and packing the remaining structures appears to
have resolved the remaining LPC hang issues on the KGPE-D16. Note that
packing the remaining structures alone was not sufficient to eliminate
the hang, however removing the packed attribute entirely (during debugging)
did resolve the hang at the expense of potential problems in ramstage.
Change-Id: If3a7509ed438870d4d05caaaaa091e1c47bf9b97
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14303
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
The SPI controller needs to be set up on devices such as the SP5100
before it can be accessed to write MCT backup data. Move the backup
data write after PCI configuration has been completed.
Change-Id: Ibcf31755242ac058407a422ce8aa33d6b0b293c7
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14305
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Always use MRC cache if possible.
Added a CRC16 array to make sure the DIMMs haven't been replaced.
In case one of the CRC's doesn't match, start normal RAM training.
Use new fallback in case of broken mrc cache.
Test system:
* Gigabyte GA-B75M-D3H
* Intel Pentium CPU G2130
Test result:
The system boots a lot faster using the MRC cache.
On swapping DIMMs the CRC16 doesn't match and normal ram training
is started.
Change-Id: Ib48fe8380446846df17d37b22968f7d4fd6b9b13
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/14172
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
This reverts commit f961becc43.
On studying the BKDG more closely this is not the correct place
to enable DIMM parity. Further patches to clarify the parity
setup process on Family 15h are forthcoming.
Change-Id: I5a3a4f1621e3048f9dfc159709410be9de6ebecd
Reviewed-on: https://review.coreboot.org/14271
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The sync flood reset fix in Change-Id: I62d897010a8120aa14b4cb8d096bc4f2edc5f248
and related changes have made it possible to move the sync flood enable statements
back into romstage.
Change-Id: I5a3a4f1621e3048f9dfc159709410be9de6ebece
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14270
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
When a fatal error and subsequent sync flood / reset occurs,
the MCA status registers may contain valuable information on
the cause of the fatal error. Add functions to report MCEs and
reset the MCA status registers early in the boot process.
Change-Id: Icde1051ac22f93688de1330f5e2c9ce28b14b59a
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14265
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Certain AMD platforms, such as those using the SP5100 southbridge,
contain a very poorly documented bug related to LPC ROM access,
which is triggered by repeated (hundreds or more) rapid calls to
get_option(). This bug manifests as a complete system deadlock
in ramstage device configuration, requiring standby power to be
removed from the system to release the deadlock.
Cache the platform ECC status to avoid repeated calls to get_option()
in the lane count detection logic.
Change-Id: I8b48c523218ccc8c113319957d6eca2d15e1070f
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14273
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
In error case die in top level function.
No functionality is changed.
Change-Id: Ie15b01184d40bdbce20d49dcab2f9fb607068c7a
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/14171
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Return errors to top level ram init function.
Required by the folowing series to implement a fallback.
No functionality is changed.
On error case the system still halts in every test.
Change-Id: I6278c4a1d7b4a96be8988a60671fc3d72cd6cb3d
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/14170
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The revision mask for all DR-* series processors was incorrectly
set to only include the DR-B revision mask. Include all DR-*
series prcessors in the DR_ALL revision mask.
Change-Id: Iceda96aa6267b24abcbf78d39f4848d2be8053b8
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Found-by: Coverity, CID 1229627 (#1 of 1): Logically dead code (DEADCODE)
Reviewed-on: https://review.coreboot.org/14216
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Enabling sync flood on DRAM MCE directly after ECC clear can
lead to a system hang with no way to determine the offending
DRAM module. Clear MCEs after ECC setup, but do not enable
sync flood until NB setup in ramstage to allow time for any
MCEs to accumulate in the status registers. Before enabling
sync flood on MCE, determine if any MCEs were logged during
ramstage execution and display them on the serial console.
Also clear the DRAM ECC sync flood bits during DRAM training
and initial ramstage execution.
Change-Id: Ibd93801be2eed06d89c8d306c14aef5558dd5a15
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14192
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
During power on from cold (S5) state, numerous MCEs are generated
before DRAM training starts, e.g. during HT link training. Clear
these MCEs before DRAM training start, and report any MCEs generated
during DRAM training.
Change-Id: I7d047571242e5bd041e4aac22c1ec1d7d26ef0e6
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14191
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
On Family 15h processors, with certain RDIMMs, MCEs are generated
as a normal part of DCT startup / DRAM training. Disable sync
flood on parity or UC data error until ECC has been enabled.
Change-Id: Ife54751ff127ffd59baaad35d3fea14ea01ef505
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14186
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
This resolves a long-standing issue with RDIMM control word
configuration failure, likely due to random parity failure.
Change-Id: If8b8dc5b8b99f4c2fe29b3a133b064631e4693be
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14184
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
In order to add a fallback mechanism, move the ram training code
into a new function. This function will be called multiple times
and must return error or success to the calling function.
Change-Id: I5ee1b3a528290d8252d236b9152b81291736958a
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/14169
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
It's required to store the dimm_info in ramctr_timing as only ramctr_timing
is written to mrc cache.
Allows to fill SMBIOS type 17 if mrc cache is used.
Change-Id: I7634b05069df307d471938d9854997a018de81b3
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/14168
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Replace open coded memset() functions with calls to the library function.
The new code also explicitly backs up and restores the data structures
that are preserved across calls to mct_ResetDataStruct_D(), and no longer
relies on structure member order to function correctly.
Change-Id: I6dd6377deda0087cd1b65f7555588978657d6516
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14165
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
During maximum read latency training on Family 15h processors,
the maximum read latency was incorrectly set from the NBP1
value instead of the correct NBP0 value.
Modify maximimum read latency training to explicitly operate
on the NBP0 value, and store the previously calculated NBP1
value for reference by other portions of the training algorithm.
Change-Id: I5d4a6c2def83df3e23f1a4c598314c31a0172cd7
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14150
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Under certain conditions (training abort) BlockRxDqsLock could
remain set in violation of the BKDG. Ensure BlockRxDqsLock is
reset to 0 after a lane training abort.
Change-Id: I1a49a24d02b2b7cacae074794ec274a424a9e66b
Reviewed-on: https://review.coreboot.org/14144
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The Intel i3100 northbridge code is the only user of
cache_ramstage(). Therefore, place it next to the sole
consumer.
Change-Id: If15fb8d84f98dce7f4de9e089ec33035622d8f74
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/14097
Tested-by: build bot (Jenkins)
Reviewed-by: Andrey Petrov <andrey.petrov@intel.com>
Reviewed-by: Furquan Shaikh <furquan@google.com>
Rebasing change I3be808db5d15ceec4c36d17582756b01425df09a
did not take into account the default UI setting introduced in
change I6ae88c891e92b21dc0ca3c47b8f3d269f83b3204 , causing DRAM
instability and occassional failure to boot.
Use the correct 1UI value for the modified function semantics.
Change-Id: I9fd24cf83e4c4083c6e467d49021c98e5f5f2c53
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/14073
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Instead of manually including udelay_io.c in each romstage,
select UDELAY_IO for all i3100 boards in the chipset.
Change-Id: Ia66a0561c75777a9e98bb87117859808a2ff3732
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/13786
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Instead of manually including udelay_io.c in each romstage,
select UDELAY_IO for all i810 boards in the chipset.
Change-Id: Ifda7dcfdf37b6affce838ee96ca6382b2d4be8c3
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/13784
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Instead of manually including udelay_io.c in each romstage,
select UDELAY_IO for all i830 boards in the chipset.
Change-Id: I0a63ddd3c5e43ea65f776385f54eceb6569751ac
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/13783
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
read_dqs_read_data_timing_registers() and
read_read_dqs_timing_control_registers() served essentially
the same function but had slightly different semantics,
causing confusion and needlessly complex Family15h code.
Consolidate both into read_dqs_read_data_timing_registers()
and adjust surrounding code to match new semantics.
Change-Id: I3be808db5d15ceec4c36d17582756b01425df09a
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13994
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The sourcecode is 99% the same. Only two lines differ, but not
in functionality.
Also rename mrccache.c -> mrc_cache.c
Tested-on: boot + suspend/resume on x220
Change-Id: I36f79d066336f223b608c70c847ea6ea6e4ad287
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: https://review.coreboot.org/14007
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The mrc_cache definition and the struct mrc_container are the same
over all intel platforms.
Change-Id: I128a4b5693d27ead709325c597ffe68a0cc78bab
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: https://review.coreboot.org/13998
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The existing MCT code proceeded to the next DRAM training phase if
the minimum lane quality standard passed for either the read or
write direction. Ensure that both pass for a given set of delay
values before proceeding to the next training phase.
Change-Id: I2316ca639f58a23cf64bea56290e9422e02edf1c
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13993
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
The AMD Family 15h BKDG rev. 3.14 indicates that the maximum read latency
must be calculated prior to DQS position training, however the read
latency calculations use read DQS delay values that have not been
set prior to DQS position training.
Set the read DQS delay values to 1UI (i.e worst case) before calculating
the read latency prior to DQS position training.
Change-Id: I6ae88c891e92b21dc0ca3c47b8f3d269f83b3204
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13995
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
A couple of arrays were not properly initialized. This
did not appear to affect operation of the codebase however
it led to some ugly values being displayed when debugging
was turned on.
Also bounds check an array index; as before this did not
appear to affect operation but was a potential point of
failure.
Change-Id: I243b7197a74aed78ddca808eb3b0f35f1fe9d95a
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13934
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Commit 71512b2c (northbridge/i945/gma: fix build error with native graphics init)
unintentionally changed the code to ignore the NVRAM setting
`tft_brightness`. Revert that hunk to restore the original behavior.
Change-Id: Iffdfc5272732bad3476f35ddac1f5a7564270531
Signed-off-by: Alexander Couzens <lynxis@fe80.eu>
Reviewed-on: https://review.coreboot.org/14002
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Instead of manually including udelay_io.c in each romstage,
select UDELAY_IO for all i440BX boards in the chipset.
Change-Id: I411191927f3fba1d0749edcf79378e8013fb195a
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/13781
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
This is a step towards isolating the timer drivers.
Change-Id: I4c9349054be0cf520cd4407be9fb393b664223a4
Signed-off-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Reviewed-on: https://review.coreboot.org/13922
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Sandy and Ivy Bridge processors use the same socket, and a mainboard
with the socket can support both types of CPUs. However, they use
different native graphics init code for LVDS and cause a crash if
running the wrong code.
This change detects the CPU type and then selects the right code to
run. It will add some more code in ramstage. It also merges the
{SANDY,IVY}BRIDGE_LVDS symbol to one SANDYBRIDGE_IVYBRIDGE_LVDS.
Tested on a Lenovo T520 with i7-2630qm and i7-3720qm
Signed-off-by: Iru Cai <mytbk920423@gmail.com>
Change-Id: I4624759f9c92d56d547db1ab4b9a1d611a182a91
Reviewed-on: https://review.coreboot.org/12087
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Tested-by: build bot (Jenkins)
Instead of hardcoding the maximum supported DDR frequency to
800Mhz (DDR3-1600), read the fuse bits that encode this information.
Test system:
* Intel IvyBridge
* Gigabyte GA-B75M-D3H
Change-Id: I515a2695a490f16aeb946bfaf3a1e860c607cba9
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13487
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The code can't handle cyclic zero runs. Make sure it will never
wrap around by setting the top-most bit to constant one.
Fixes "Mini channel test failed (2)".
Test system:
* Gigabyte GA-B75M-D3H
* Intel Pentium CPU G2130
Change-Id: I55e610d984d564bd4675f9318dead6d6c1e288a3
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13853
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
I first found the missing of #include guards when I tried to include
both sandybridge/gma.h and sandybridge/sandybridge.h, but
sandybridge.h includes gma.h in it and gives a compile error.
Change-Id: I13fdb8014b82e6065be2064137b7ea10062deaca
Signed-off-by: Iru Cai <mytbk920423@gmail.com>
Reviewed-on: https://review.coreboot.org/13775
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Improved version of
I1a115a45d5febf351d89721ece79eaf43f7ee8a0
The first version wasn't well tested due to the lack of hardware
and it was to aggressive.
With timC being direct function of timB's 6 LSBs it's critical to match
timC and timB.
Some tests increments the value of timB by a small value,
which might cause the 6bit value to overflow, if it's close
to 0x3F.
Increment the value by a small offset if it's likely
to overflow, to make sure it won't overflow while running
tests and bricks the system due to a non matching timC.
In comparission to the first attempt, only 4 out of 128 timB values
are considered bad.
Needs test on real hardware !
Fixes a "edge write discovery failed" on my test system.
Test system:
* Intel IvyBridge
* Gigabyte GA-B75M-D3H
Change-Id: If9abfc5f92e20a8f39c6f50cc709ca1cedf6827d
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13714
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Some vendors store lower frequency profiles in the regular SPD,
if the SPD contains a XMP profile. To make use of the board's and DIMM's
maximum supported DRAM frequency, try to parse the XMP profile and
use it instead.
Validate the XMP profile to make sure that the installed DIMM count
per channel is supported and the requested voltage is supported.
To reduce complexity only XMP Profile 1 is read.
Allows my DRAM to run at 800Mhz instead of 666Mhz as encoded in the
default SPD.
Test system:
* Gigabyte GA-B75M-D3H
* Intel Pentium CPU G2130
Change-Id: Ib4dd68debfdcfdce138e813ad5b0e8e2ce3a40b2
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13486
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: build bot (Jenkins)
Use printram() in more places and use printk() only where
it makes sense.
Remove spamming "MRd: %x <= %x\n".
Use common syntax for timing output.
Change-Id: I38965967a029994112d7ab63afd4d9968a7728c5
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13414
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Issue observed:
The PCIe Root port shows up in GNU/Linux but no PCIe device
is being detected.
Test system:
* Gigabyte GA-B75M-D3H (Intel Pentium CPU G2130)
* Lenovo T530 (Intel Core i5-3320M CPU)
Problem description:
The PEG Root port link training on Ivy Bridge needs to be manually started.
Problem solution:
The bits are set in early_init to meet PCIe reset timeout of 100msec.
The bits should be set in PCI device enable function, but this causes the
PCI enumeration to not detect the card, as it's still booting. Adding
a fixed delay of 100msec resolves this problem, but this would
increase boot time.
Read the PCI base revision mask to make sure it's any IvyBridge CPU.
Don't run the code on MRC path as it has its own PEG initilization code.
Tested with:
* Nvidia NVS 5400M (PCIe2)
* ATI Radeon HD4780 (PCIe2)
* Nvidia GeForce 8600 GT (PCIe1)
Untested:
* PCIe3 devices
Final test results:
The PEG device shows up under GNU/Linux and can be used without issues.
Change-Id: Id8cfc43e5c4630b0ac217d98bb857c3308e6015b
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/11917
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Use shared gpio code from common folder.
Bd82x6x's gpio.c and gpio.h is used by other southbridges
as well and will be removed once it is unused.
Change-Id: I8bd981c4696c174152cf41caefa6c083650d283a
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13614
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
It looks like the falling timing was missing the shift offset.
Not sure if this was intentional, I guess not.
Tested on my hardware and produced no regressions.
Test system:
* Intel IvyBridge
* Gigabyte GA-B75M-D3H
Please test on real hardware !
Change-Id: Id8c60217093a48bf322f406ea258c10a02c936e8
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13682
Tested-by: build bot (Jenkins)
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Otherwise the image is simply unusable.
Change-Id: I1e2562ba17279d14dc73b05e4f8fa493e06fbcd2
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: https://review.coreboot.org/13699
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
We want the question for CBFS size to be next to the rom size in the
mainboard directory, but that doesn't seem to work for how people
want to set the defaults. Instead of having the list of exceptions
to the size, just set the defaults at the end of kconfig.
- Move the defaults for chipsets not setting HAVE_INTEL_FIRMWARE into
the chipset Kconfigs (gm45, nehalem, sandybridge, x4x)
- Override the default for HAVE_INTEL_FIRMWARE on skylake.
- Move the HAVE_INTEL_FIRMWARE default setting into the firmware
Kconfig file
- Move the location of the default CBFS_SIZE=ROM_SIZE to the end of
the top level kconfig file, while leaving the question where it is.
Test=rebuild Kconfig files before and after the change, verify that
they are how they were intended to be.
Note: the Skylake boards actually changed value, because they were
picking up the 0x100000 from HAVE_INTEL_FIRMWARE instead of the
0x200000 desired. This was due to the SOC_INTEL_SKYLAKE being after
the HAVE_INTEL_FIRMWARE default. Affected boards were:
Google chell, glados, & lars and Intel kunimitsu.
Change-Id: I2963a7a7eab037955558d401f5573533674a664f
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/13645
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In the same time remove few native gfx options which were improperly set
and only added dead code to the binary.
Change-Id: I4ed3fec03a1655ae0a779c3aa3845de273cb12e1
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: https://review.coreboot.org/13649
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: build bot (Jenkins)
This is needed for stout EC init.
Change-Id: I5c73499c17763229840152a473a2d820802ee2f6
Signed-off-by: Vladimir Serbinenko <phcoder@gmail.com>
Reviewed-on: https://review.coreboot.org/13535
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
This reverts commit 0e06f5bd70.
It breaks gm45 and also does some magic without being asked too. It
disables bridge devices permanently if no device was found on the se-
condary bus. In a simple notebook world this might be ok, but it breaks
hot-plugging and late detection (if a secondary bus device comes up too
slow for the firmware to detect and the OS has to enumerate it).
Change-Id: Ia2010640d7c55b0bdd44164b81c75dd4be50410b
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/13609
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Certain registered DIMMs failed training due to an error
likely introduced during historical rebase. Ensure that
the SubMemclkRegDly bit is set according to BKDG
recommendations on Family 15 processors.
Change-Id: I24c95265dada9eabf4df280b6f2b4a1eb9cecaf1
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13148
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Under certain conditions, not elucidated in the BKDG,
an extra memclock of CAS write latency is required.
The only reliable way I have found to detect when this
is required is to try training without the delay, and
if DQS position training fails, adding the delay and
retraining.
This is probably related in some form or another to
the badly broken DQS Write Early algorithm given
in the BKDG.
Change-Id: Idfaca1b3da3f45793d210980e952ccdfc9ba1410
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13531
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Walk the bus and try to find enabled devices.
Disable the PEG port if no devices are attached.
Change-Id: I67fcc831fd78ecc6dba83f4e0662ec7549cc2591
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/12894
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Issue observed:
The system boots with 4G in channel 0 and 4G in channel 1.
The system doesn't boot with any combination of 4G + 1G in
channel 0 and 4G in channel 1.
In both cases DIMM1 failed, while DIMM0 showed no issues.
Problem description:
The CLK to CMD/CTL was off by a half clock cycle.
The find the issue I X-Y plotted timC vs timB for every
lane on the failing rank.
You can see an offset by 32 units for timB, that is not present on
other ranks.
It turns out that the XOVER CMD/XOVER CTL enable bit for DIMM1 was
missing in program_timings(), which caused the clock offset.
Problem solution:
Add two functions to calculate XOVER CMD and XOVER CTL and use both
to set XOVER in program_timings() and dram_xover().
Final testing result:
The system boots with 4G + 1G in channel 0 and 4G in channel 1.
Test system:
* Intel IvyBridge
* Gigabyte GA-B75M-D3H
Change-Id: I88694c86054ade77e9d8bb2f1fdaf7bc559c1218
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/13415
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
This fixes some spelling and whitespace issues that I came across
while working on various things in the tree.
There are no functional changes.
Change-Id: I33bc77282f2f94a1fc5f1bc713e44f72db20c1ab
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/13016
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
On certain Winbond SuperIO devices, when a PS/2 mouse is not
present on the auxiliary channel both channels will cease to
function if the auxiliary channel is probed while the primary
channel is active. Therefore, knowledge of mouse presence
must be gathered by coreboot during early boot, and used to
enable or disable the auxiliary PS/2 port before control is
passed to the operating system.
Add auxiliary channel PS/2 device presence detect, and update
the Winbond W83667HG-A driver to flag the auxiliary channel as
disabled if no device was detected.
Change-Id: I76274493dacc9016ac6d0dff8548d1dc931c6266
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13165
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The existing code used an incorrect macro name to check for mainboard
DRAM voltage set support, and as a result no voltages were actually
set. Furthermore, the existing code did not contain a centralized
voltage assumption for boards that did not have a DIMM voltage set
implementation.
Use the correct macro name to test for boards with voltage set
implementation, and provide a basic fallback to 1.5V operation
for boards without a voltage set implementation.
Change-Id: I638c65fe013a8e600694d8cbedf6a10b33b0ef95
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13150
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The existing code generated an incorrect boot APIC ID from node and
core number for single node packages, leading to a boot failure when
the second node was installed.
Properly generate the boot APIC ID from node and core number.
Change-Id: I7a00e216a6841c527b0a016fa07befb42162414a
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13149
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Martin Roth <martinroth@google.com>
Previously with errors in the ram init, early cbmem was disabled.
Now that the ram is working correctly, set as early cbmem platform
and update all (1) boards to use it.
Tested on GA-G41M-ES2L
Change-Id: I5925c28821537f0e326b4f5a2ac39778e4724a3c
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/13131
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tidy up the code and move vga_textmode_init() later
Change-Id: I49967e7197416c955ae6c8775eac7d1a60c92d1c
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/13128
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
- Fix bug with msbpos, it was not returning the correct result
due to typo in logic, and unsigned value needed to be negative.
- Add reclaim above 4GiB
- Fix to ME related registers near the end of raminit
Change-Id: I04acd0593a457437ee4a42e14b287b2b17a160af
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/13127
Tested-by: build bot (Jenkins)
Reviewed-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Previously, 0xa0000000 to 0xc0000000 needed to be reserved as
a non-usable memory hole because it would hang on memory i/o.
Memtest86+ now passes with no errors on both channels populated.
Tested on GA-G41M-ES2L with 2x2GiB sticks of ram.
Change-Id: Ib52a63a80f5f69c16841f10ddb896ab3c7d30462
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/13125
Reviewed-by: Aaron Durbin <adurbin@chromium.org>
Tested-by: build bot (Jenkins)
This eliminates all "ud2" instances from romstage disassembly.
Change-Id: I3b0c8322a4ca4a851b0cce8f3941425d9cb30383
Signed-off-by: Patrick Georgi <patrick@georgi-clan.de>
Reviewed-on: https://review.coreboot.org/13488
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
VGA grub console works but display wobbles left/right
drm/i915 driver reports one error:
- [drm:i915_irq_handler] *ERROR* pipe A underrun
- Monitor does not display 1920x1080 after modeset
- Other resolutions look out of sync
Cause: suspect single bug in raminit (chipset init)
Change-Id: I2dcf59f8f30efe98f17a937bf98f5ab7221fc3ac
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/12921
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The existing code accidentally truncated the MSB from the MR0
WR value. While this probably had a minimal effect in reality,
it should be configured correctly for maximal system stability.
Change-Id: Ifb8a39c6ca47b32b44d33735e5c6c39f1dc5a44e
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13147
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
The existing drive strength calibration code did not strictly
follow the BKDG-defined setup process. Bring the calibration
code in line with the BKDG recommendations.
Change-Id: I122eeb93958d88de59d0c3b2979f607afa2c52c3
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13145
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
When an Extended Temperature Range DIMM is installed on a channel
the refresh rate should be increased per the BKDG recommendations
to allow correct operation at higher temperature ranges.
Set fast refresh on a channel if an ETR DIMM is installed on that
channel.
Change-Id: I7a085d34efc78f3f0794a5cb33b88f27a5e6d54e
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13144
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The existing MCT initialization code was largely missing C32 socket-
specific configuration data. Add C32 socket-specific timing and ODT
values as specified in the BKDG.
Change-Id: I8eef8d5c8581f03d269663a338d5542744c5cdd7
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13141
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The existing code applied G34-specific speed limits to all socket
types. Update G34 and C32 specific speed limits to be in line with
BKDG recommendations.
Change-Id: I958ad333c47948ae741a56de5866af3e636fd24d
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/13140
Tested-by: build bot (Jenkins)
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Fixes bug that decode_pciebar() function was bypassed due
to PCI_DEV(0,0,0) being detected as zero and function returning 0.
Change-Id: Ia79bcebbe3ba36f479cbb24dbbb163a031d9c099
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/13031
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-by: Vladimir Serbinenko <phcoder@gmail.com>
This just updates existing guard name comments on the header files
to match the actual #define name.
As a side effect, if there was no newline at the end of these files,
one was added.
Change-Id: Ia2cd8057f2b1ceb0fa1b946e85e0c16a327a04d7
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12900
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Count DIMMs on current memory channel instead of all memory channels.
The current code is only able to correctly handle the following memory
configurations:
One DIMM installed in either channel.
Four DIMMs installed, two in each channel.
Two DIMMs installed, both in the same channel.
For systems that have any other configuration the DRAM On-Die-Termination
setting is wrong.
For example:
Two DIMMs installed, one in each channel.
Test system:
* Gigabyte GA-B75M-D3H (Intel Pentium CPU G2130)
Change-Id: I0e8e1a47a2c33a326926c6aac1ec4d8ffaf57bb6
Signed-off-by: Patrick Rudolph <siro@das-labor.org>
Reviewed-on: https://review.coreboot.org/12892
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Philipp Deppenwiese <zaolin.daisuki@googlemail.com>
Register settings are the same as on newer chips (compare sandy-
bridge), just at different locations.
Change-Id: Iea0359165074298a376e0e2ca8f37f71b83ac335
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/12885
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
Skip everything but the final setting of PP_CONTROL, i.e. triggering
the power up. The settings with PANEL_UNLOCK_REGS are useless as no
lockable registers were touched in between. Also the loop waiting for
the panel power up to finish was a no-op as the registers with the
power timings were never filled (see follow-up commits).
Change-Id: Ife27dcafdf197b2246c4e69f2bf7a3a6765d1d82
Signed-off-by: Nico Huber <nico.h@gmx.de>
Reviewed-on: https://review.coreboot.org/12884
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This continues what was done in commit a73b93157f
(tree: drop last paragraph of GPL copyright header)
Change-Id: Ifb8d2d13f7787657445817bdde8dc15df375e173
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12914
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
- Don't redefine D0F0_PCIEXBAR_LO, use the #define in x4x.h
- Move TPMBASE and TPM32() definitions into iomap.h
- Use "" style include for x4x.h in nortbridge files.
- Move includes of .h files out of x4x.h and into the c files that need
them.
- Protect function definitions in bootblock.
Change-Id: I3fdb579235c5446733a0ffba05fffe1a73381251
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12849
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
This reverts commit d3deecdd9c.
Do not mix open-source AGESA and binary PI trees. Once you have
working S3 support for binaryPI platforms, add the adapted
oem_s3.c file as northbridge/amd/pi/oem_s3.c instead.
Change-Id: I7c981d0023a5c0225e046f9c0104acfa07436b79
Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-on: https://review.coreboot.org/12282
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Make the low-power and small form factor (SFF) options overridable
from romstage main. Also disable both options by default. That's ok
as there aren't yet any in-tree users of the GS45 chipset. As a nice
side-effect, this adds X200s support to the lenovo/x200 port.
Change-Id: I94373851262e6d424cf4885ceca7260c31bc9f61
Signed-off-by: Nico Huber <nico.huber@secunet.com>
Reviewed-on: https://review.coreboot.org/12814
Tested-by: build bot (Jenkins)
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Martin Roth <martinroth@google.com>
Passes memtest86+ with either one or two sticks of 2GB ram
but memory map needs a hole at 0xa0000000 to 0xc0000000
Change-Id: Ib34d862cb48b49c054a505fffcba1c17aeb39436
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/11307
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Non-code flow assembly stubs do not have to be included in
bootblock.S, now that we have more freedom in bootblock linking.
Rather than bringing these stubs to the config system, just link them
in the bootblock.
Note that we cannot fully remove CHIPSET_BOOTBLOCK_INCLUDE at this
point, as some intel SOCs use this stub for code flow.
objdump -h build/cbfs/fallback/bootblock.debug on a few random boards
confirms that the appropriate sections are still included in the
final binary.
Change-Id: Id3f9ece14e399c1cc83090f407780c4a05a076f0
Signed-off-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Reviewed-on: https://review.coreboot.org/11856
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Boots to console on Gigabyte GA-G41M-ES2L
Ram initialization *not* included in this patch
VGA native init works on analog connector
Change-Id: I5262f73fd03d5e5c12e9f11d027bdfbbf0ddde82
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/11305
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
As the southbridge largely controls the PCI[e] configuration space
this patch moves the resource allocation from the northbridge
to the southbridge when the extended configuration space region
is enabled.
Change-Id: I0c4ba74ddcc727cd92b848d5d3240e6f9f392101
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12050
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
The remainder of the divide operation was being placed into a Local,
but was never being used, causing an IASL warning. Since this
field is optional, just remove the Local.
Fixes IASL warning:
dsdt.aml 640:Divide (Multiply (CTDN, 125), 100, Local0, PL2V)
Warning 3144 - Method Local is set but never used ^ (Local0)
Change-Id: I0b43ef638b1bc3e1163c45f31f8da57aa0d39e22
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12706
Tested-by: build bot (Jenkins)
Reviewed-by: Marc Jones <marc@marcjonesconsulting.com>
The ALIGN_CURRENT macro relied on a local variable name
as well as being defined in numerous compilation units.
Replace those instances with an acpi_align_current()
inline function.
Change-Id: Iab453f2eda1addefad8a1c37d265f917bd803202
Signed-off-by: Aaron Durbin <adurbin@chromium.org>
Reviewed-on: https://review.coreboot.org/12707
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The Kconfig symbols were missing an underscore, so were not getting
evaluated properly.
Change-Id: I619cf3f44f44f9c9699482d64164d3db28cd4c8f
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12559
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
In coreboot, bool, hex, and int type symbols are ALWAYS defined.
Change-Id: I58a36b37075988bb5ff67ac692c7d93c145b0dbc
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12560
Tested-by: build bot (Jenkins)
Reviewed-by: Stefan Reinauer <stefan.reinauer@coreboot.org>
The PER_DEVICE_ACPI_TABLES Kconfig symbol is no longer used as it was
removed in commit 83f81cad (acpi: Remove monolithic ACPI)
Change-Id: Ie6ba252f6e7d33da9d4500f1201367f116e4c505
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12554
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Does native ram init for Intel Atom D5xx 8086:a000 northbridge
Tested on Intel D510MO mainboard, board boots linux kernel
- Works fully with both dimms populated (2x2GB), memtest passes 100%
- Almost boots with only one dimm in one of the slots
(suspect bad memory map with one dimm?)
- Reads garbage with only one dimm in other slot
Change-Id: Ibd22be2a959045e0a83aae2a3a0e877013f80711
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: https://review.coreboot.org/12501
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The existing code re-used the Family 10h XCS buffer setup on
Family 15h CPUs, which set incorrect values leading to random
system lockups.
Use the Family 15h XCS buffer setup shown in the BKDG.
Change-Id: Ie4bc8b3ea6b110bc507beda025de53d828118f55
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12070
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The secondary bus number set code incorrectly overwrote the link
buffer settings in F0x[F4,D4,B4,94]. Constrain the secondary
bus number set to the appropriate bits of the registers.
Change-Id: If70825449f298aa66f7f8b76dbd7367455a6deb1
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12068
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The Family 15h DRAM initialization did not set up the various
tristate enable codes in the MCT.
Add Family 15h tristate enable setup. This fixes multiple
DIMMs on a single channel.
Change-Id: I0278656e98461882d0a64519dfde54a6cf28ab0f
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12060
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Tested-by: Raptor Engineering Automated Test Stand <noreply@raptorengineeringinc.com>
The Kconfig symbol CONFIG_HT3_SUPPORT is not implemented.
This mirrors commit c5163ed8 (AMD binaryPI: Drop HT3_SUPPORT)
Change-Id: I2682d3b620e2cee613c7421622a8c79db5ba3a86
Signed-off-by: Martin Roth <martinroth@google.com>
Reviewed-on: https://review.coreboot.org/12556
Tested-by: build bot (Jenkins)
Reviewed-by: Kyösti Mälkki <kyosti.malkki@gmail.com>
There is no need to continue testing a DCT configuration after
data errors have already been detected; this just wastes time
during boot.
Change-Id: I979e27c32a3e0b101590fba0de3d7a25d6fc44d2
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12066
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
The BKDG recommends the use of an antiphase window detection
algorithm to ensure that the DQS data eye is properly centered.
TEST: Booted both with DIMMs known to move the data eye into the
prior clock phase and DIMMs known to keep the data eye in the
current clock phase.
Change-Id: I1d85fddd45197ca82dcaa46fe863e64589712d1f
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12059
Tested-by: build bot (Jenkins)
Reviewed-by: Edward O'Callaghan <edward.ocallaghan@koparo.com>
Reviewed-by: Martin Roth <martinroth@google.com>
The odd rank of each DIMM could experience data corruption due to
incorrect DQS training. Fix the DQS training algorithm by executing
the relevant portions of the training algorithm on the odd ranks.
Change-Id: Ibc51f5052d5189e45b3d9aa98ca8febbfe13f178
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12058
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
Also fix incorrect Trfc[0-3] value on Family 15h.
Change-Id: Iafc233984ae1d44fe6a1cb5b109d36397cbd991a
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12055
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The silicon in control of CC6 appears to contain minor bugs
and / or deviations from the BKDG; through trial and error
it was found that these issues can be worked around by reserving
the entire possible CC6 save region, regardless of currently
installed node count.
Change-Id: If31140651f25f9c524a824b2da552ce3690eae18
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: https://review.coreboot.org/12054
Tested-by: build bot (Jenkins)
Reviewed-by: Patrick Georgi <pgeorgi@google.com>
The coherent fabric on all Family 10h/15h devices supports
isochronous mode, which is required for IOMMU operation.
Add initial support for isochronous operation.
Change-Id: Idd7c9b94a65f856b0059e1d45f8719d9475771b6
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12042
Tested-by: build bot (Jenkins)
Reviewed-by: Martin Roth <martinroth@google.com>
Based on i945. Tested on Intel D510MO mainboard,
board boots to UART console with this code.
Change-Id: I1d92a1aa6d6d767bda8379807dc26b50b9de75c9
Signed-off-by: Damien Zammit <damien@zamaudio.com>
Reviewed-on: http://review.coreboot.org/10073
Tested-by: build bot (Jenkins)
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Certain workloads may evict too many lines of other cores from the
L3 cache if configured as one monolithic shared cache region.
Forcibly partition L3 cache to improve performance.
Change-Id: Ie4e28dd886aaa1c586b0919c5fe87ef1696f47e9
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12036
Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net>
Tested-by: build bot (Jenkins)
Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
The existing code generated an invalid NUMA table
that was rejected by Linux, leading to poor resource
allocation. This was due to system MMIO resources
being inserted into the table when the table should
only contain DRAM resources.
Do not include system MMIO resources (i.e. resources
with an index less than 0x10) in the NUMA table.
Change-Id: I99c200382b52a99687daf266a84873d9ae2df025
Signed-off-by: Timothy Pearson <tpearson@raptorengineeringinc.com>
Reviewed-on: http://review.coreboot.org/12035
Reviewed-by: Alexandru Gagniuc <mr.nuke.me@gmail.com>
Tested-by: build bot (Jenkins)
