summaryrefslogtreecommitdiff
path: root/src/soc/amd
AgeCommit message (Collapse)Author
2021-04-30amd/cezanne: Add telemetry setting to UPDChris Wang
Add telemetry setting to UPD, the value comes from the SDLE testing. BUG=b:182754399 TEST=Build & Boot guybrush Cq-Depend: chrome-internal:3787638 Signed-off-by: Chris Wang <chris.wang@amd.corp-partner.google.com> Change-Id: I9dd3643e9c582a41192130901935eef321b2c67e Reviewed-on: https://review.coreboot.org/c/coreboot/+/52733 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-29soc/amd/common: Remove eSPI decode workaroundRaul E Rangel
We no longer lock up if we clear the port 80 bit. I'm assuming this was fixed when we configured the PSP to no longer setup eSPI. BUG=b:183974365 TEST=Boot guybrush to OS Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I1530d08974d42e0b06eb783521dea32fca752d85 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52677 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Martin Roth <martinroth@google.com>
2021-04-29psp_verstage: make temp_stack optionalKangheui Won
Temp stack for verstage is only needed for picasso, so make it optional in the layout file. Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I44196103a3531e9d01c96ab8f454c8b580fe9807 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52688 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Martin Roth <martinroth@google.com>
2021-04-29psp_verstage: make get_max_workbuf_size optionalKangheui Won
From cezanne we have enough space in PSP so we don't have to worry about workbuf size. Hence the function only exists in picasso and deprecated for later platforms. So wrap svc_get_max_workbuf_size and provide default weak function so future platforms don't have to implement dumb function for it. TEST=build and boot zork, check weak function is not called in zork Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I16e8edf8070aaacb3a6a6a8adc92b44a230c3139 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52687 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Martin Roth <martinroth@google.com>
2021-04-29soc/amd/picasso: move PSP_SRAM addrs to separate headerKangheui Won
These addresses will be changed in cezanne. Before start working on cezanne, move these out to separate header as a clean-up. TEST=emerge-zork coreboot Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I2499281d250aae701f86bfcc87c7681e5b684b6a Reviewed-on: https://review.coreboot.org/c/coreboot/+/52625 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-29soc/amd/cezanne: Enable Audio Co-processor driverKarthikeyan Ramasubramanian
BUG=b:182960979 TEST=Build and boot to OS in Guybrush. Change-Id: I73d1d3e5c1c4eb30ebf44f38d381beba84075351 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52644 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-29soc/amd/common/acp: Move Audio Co-processor driver to commonKarthikeyan Ramasubramanian
Audio Co-processor driver is similar for both Picasso and Cezanne SoCs. Hence move it to the common location. BUG=None. TEST=Builds Dalboz, Trembyle, Vilboz, Mandolin and Bilby mainboards. Change-Id: I91470ff68d1c183df9a2927d71b03371b535186a Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52643 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-28soc/amd/common/smi_handler: Print warning when receiving an SCI SMIRaul E Rangel
We don't have any infrastructure setup to handle SCI SMIs. Instead of just silently ignoring the SMI, print a warning saying that it is being ignored. BUG=none TEST=Trigger an SCI SMI and see warning printed. Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I803e572250925b7d5ffdbb3e8958f9aff1f808df Reviewed-on: https://review.coreboot.org/c/coreboot/+/52674 Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28soc/amd/cezanne: copy psp_transfer.h from picassoKangheui Won
Cezanne version of psp_transfer.h lacks some necessary definitions. Currently we don't have any plan to change transfer buffer structure in cezanne, so just copy'em over. Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I9361c4ab76c8ded06358a7718d5e447c16414721 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52540 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-28soc/amd/cezanne: copy Kconfig options for psp_verstageKangheui Won
These are just copied from picasso one. Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I701d6af63b24e86f8e132fad73504e20148a2bf3 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52539 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2021-04-26soc/amd/cezanne: Update STAPM vars with unitsMartin Roth
Like the Picasso platform, it's very useful to have units on these variables. BUG=b:185209734 TEST=Build & Boot Signed-off-by: Martin Roth <martinroth@chromium.org> Change-Id: I592c807c5e9a2c17b1c5959e56a01237352c5204 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52649 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Eric Peers <epeers@google.com> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-26amd/cezanne: Add slow_ppt_time & thermctl_limit to UPDMartin Roth
These values will be added in the upcoming STAPM configuration update. BUG=b:185209734 TEST=Build & Boot guybrush Cq-Depend: chrome-internal:3780259 Signed-off-by: Martin Roth <martinroth@chromium.org> Change-Id: I2a6835c16badfe505e3c33b356ca671766cd6972 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52648 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Eric Peers <epeers@google.com> Reviewed-by: chris wang <Chris.Wang@amd.com> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-26src/soc/amd/picasso: Add HDMI 2.0 disable settingPatrick Huang
hdmi2_disable bit0~3 is used to disable HDMI 2.0 function in DDI0~3 BUG=b:179170193 BRANCH=none TEST=Build; Verify the UPD was passed to system integrated table Signed-off-by: Patrick Huang <patrick.huang@amd.corp-partner.google.com> Change-Id: I383bfd04e01f5202db093105662344869e475746 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52545 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Martin Roth <martinroth@google.com> Reviewed-by: Nikolai Vyssotski <nikolai.vyssotski@amd.corp-partner.google.com>
2021-04-26soc/amd/cezanne/fsp_m_params: Configure the s0i3_enable UPDKarthikeyan Ramasubramanian
Configure the S0i3 enable UPD based on the mainboard configuration. BUG=b:178728116 TEST=Build and boot to OS in Guybrush. Enter S0i3 after passing the sleep state configuration from the mainboard. Change-Id: I18f43e964d1c70317155394257a5e2c1900816bb Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52618 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Jason Glenesk <jason.glenesk@amd.corp-partner.google.com> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-26soc/amd/cezanne,picasso/Kconfig: add help text for MAX_CPUSFelix Held
Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Suggested-by: Nico Huber <nico.h@gmx.de> Change-Id: I76270b43b3202bda71ff3f6b97d5ffa2234511b7 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52646 Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-26soc/amd/cezanne & picasso: Add Kconfig for hardcoded Soft Fuse bitsMartin Roth
Currently, some of the PSP Soft Fuse bits are hardcoded in the Cezanne and Picasso makefiles. This makes it impossible for platforms to change them. This change puts the hardcoded bits in Kconfig, allowing them to be modified by the platform. BUG=b:185514903 TEST=Verify that the correct Soft Fuse bits are set. Signed-off-by: Martin Roth <martinroth@chromium.org> Change-Id: I190ebf47cb7ae46983733dc6541776bf19a2382f Reviewed-on: https://review.coreboot.org/c/coreboot/+/52422 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-23soc/amd/picasso/mca: fix CTL_MASK MSR accessFelix Held
MC0_CTL_MASK is no longer available in fam 17h and newer and will result in a general protection fault when accessed. This register was moved, so use the one that is correct for this CPU generation. BUG=b:186038401 TEST=Mandolin no longer crashes in the machine check error handling path with a general protection fault. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Ibb042635d917dfcb2121849e2913aa62eca09dd0 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52583 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-23soc/amd/common/block/include/amdblocks: add msr_zen.hFelix Held
Add defines for the Machine Check Architecture Extensions (MCAX) MSRs and the new MCA_CTL_MASK MSRs. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Id78483e7df00c3e99c698c0344f38be68d1dfb72 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52615 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-23amd/common/blocks: Print eSPI peripheral channelRaul E Rangel
BUG=none TEST=Boot guybrush with ESPI_DEBUG Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I4312aaedcfed1535ef00a4686f218d30e351b33f Reviewed-on: https://review.coreboot.org/c/coreboot/+/52226 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-23soc/amd/picasso/cpu: make one line comment use only one lineFelix Held
Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I264e44132a6a9df6f548c9856c2256d1b92916c1 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52612 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-23soc/amd/picasso/cpu: make sure that MAX_CPUS isn't overriddenFelix Held
Trying to limit the number of available cores by setting the MAX_CPUS Kconfig option to a lower value than the SoC's default might result in cores being enabled in the FSP-S, but not fully initialized in coreboot which will cause some malfunction. Add a static assert to make sure that this option isn't changed from the default. To limit the maximum number of cores, use the downcore_mode and disable_smt devicetree settings instead. TEST=Build fails if MAX_CPUS isn't the expected default. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I3cfe09f8bb89a2154d37a37398df982828c824f9 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52611 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-23soc/amd/cezanne/cpu: make sure that MAX_CPUS isn't overriddenFelix Held
Trying to limit the number of available cores by setting the MAX_CPUS Kconfig option to a lower value than the SoC's default might result in cores being enabled in the FSP-S, but not fully initialized in coreboot which will cause some malfunction. Add a static assert to make sure that this option isn't changed from the default. To limit the maximum number of cores, use the downcore_mode and disable_smt devicetree settings instead. BUG=b:184162768 TEST=Build fails if MAX_CPUS isn't the expected default. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Idd6aa1d99128b17218a8e910c33415218a58578f Reviewed-on: https://review.coreboot.org/c/coreboot/+/52606 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-23soc/amd/picasso: clean up Kconfig and headerKangheui Won
Clean up Kconfig and psp_trasfer.h files before copying over to cezanne. TEST=build, flash and boot on jelboz360 Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: Ib370d93e23c15a2fe4c46051ed3647d2d067bb10 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52563 Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-23soc/amd/cezanne: fix i2c compiler errors on non-x86Kangheui Won
if ENV_X86 is not true we had several compile errors in i2c code. Fix them before we add code for psp_verstage which is non-x86. BUG=b:182477057 BRANCH=none TEST=build Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I0796671dd34ab2d0f123c904a88c57cdad116a57 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52538 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2021-04-23amd/vboot: remove bl_syscall_public.h from includeKangheui Won
bl_syscall_public.h is a header file for PSP app, but was used for x86 code to get the definition of PSP_INFO. Move the definition into psp_transfer.h and do not include bl_syscall_public.h from x86 code. BUG=none TEST=build psp_verstage on zork BRANCH=none Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I0fe011652a47d0ba2939dc31ee3b83f0718a61dc Reviewed-on: https://review.coreboot.org/c/coreboot/+/52537 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Martin Roth <martinroth@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-22soc/amd/common/block/cpu/noncar/memmap: include types.hFelix Held
size_t is used in the code, so we should include types.h instead of stdint.h to also have those type definitions. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I04c50de257a2d6982bfd4907eb5a1325a751919a Reviewed-on: https://review.coreboot.org/c/coreboot/+/52582 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-22guybrush: Add Kconfig for PSP eSPI and port80Rob Barnes
Add PSP_DISABLE_POSTCODES and PSP_POSTCODES_ON_ESPI kconfig options for cezanne. Select PSP_DISABLE_DISABLE_POSTCODES and unselect PSP_POSTCODES_ON_ESPI for guybrush. Port80 codes from PSP can cause bus errors on guybrush. BUG=b:185514903, b:184356693 TEST=Boot guybrush, observe no port80 codes from PSP Change-Id: I7241e47ec1b89782e699135370c796eb251afcaa Signed-off-by: Rob Barnes <robbarnes@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52401 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/{cezanne,common}/acpi: Add _OSC methodRaul E Rangel
The linux kernel requires a valid _OSC method. Otherwise the _LPI table is ignored. See https://source.chromium.org/chromiumos/chromiumos/codesearch/+/main:src/third_party/kernel/v5.10/drivers/acpi/bus.c;l=324 Before this patch: acpi_processor_get_lpi_info: LPI is not supported After this patch: acpi_processor_evaluate_lpi: ACPI: \_SB_.CP00: ACPI: Found 4 power states BUG=b:178728116 TEST=Boot OS and verify _LPI table is parsed Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I44e554b9db6f70fdd1559105cdaee53aeb2bfbf5 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52528 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-21soc/amd/cezanne: add SMU settings to devicetreeFelix Held
BUG=b:182297189 TEST=none Cq-Depend: chrome-internal:3772425 Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Ifbcc85cc10d59f1418bbf0ed4a0dc7549d589a26 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52198 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-21soc/amd/picasso/chip.h: improve comments on downcore_modeFelix Held
Clarify that the downcoring is about deactivating physical cores. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Ib8a9d1cedff995c507c3be72e7665953e1659238 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52554 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/picasso/chip.h: use boolean type for smt_disableFelix Held
Even though the UPD field this information is finally written to is an 8 bit value, the smt_disable option is only a boolean. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Iaac49944993a28ffb98a80201effe1238ec60875 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52553 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/picasso/chip.h: use types.hFelix Held
Since the next patch will use a boolean, replace the stddef.h and stdint.h includes with types.h to have all that we'll need. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I0d062c8de29aa3688a911d7887faf592020b33c2 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52552 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/cezanne: add downcoring and SMT disable settings to devicetreeFelix Held
BUG=b:184162768 TEST=none Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Id03454ed2be242bce9497560c089f75046ed7e32 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52197 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/cezanne/chip.h: include missing types.hFelix Held
Since we use uintX_t, bool and friends, we need to make sure to include the corresponding definitions. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Icb8a6e93d7f1923ac95e584fb3e33c391963f5ab Reviewed-on: https://review.coreboot.org/c/coreboot/+/52551 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/cezanne: Add support for C-state 3Raul E Rangel
These values match the majolica UEFI firmware. BUG=b:185787242, b:178728116, b:185921043 Signed-off-by: Jason Glenesk <jason.glenesk@amd.corp-partner.google.com> Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: If107c7e836942eeba734c1634fa7f8555c3018b1 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52526 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-21soc/amd/picasso/acp: use clrsetbits32 in acp_update32Felix Held
Use existing functionality instead of reinventing it. TEST=none Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Iaeab5cce05ccd860bc8de3775b7d1420653497a3 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52525 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-21soc/amd/picasso/acp: rename acp_update32 mask parametersFelix Held
The name of the and_mask parameter was a bit misleading, due to the function inverting the value. Renaming this into clear and set makes it more obvious what those parameters will actually do. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: If307ab4858541861e22f8ff24ed178d47ba70fe5 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52524 Reviewed-by: Angel Pons <th3fanbus@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-20soc/amd/{common,cezanne}: Add uPEP deviceRaul E Rangel
The uPEP device is required to support S0i3. The device has been written in ASL to make it easier to read and maintain. The device constraints are purely informational. We use a dummy constraint like the Intel platforms to keep both linux and Windows functional. In order for this device to be used by the linux kernel the ACPI_FADT_LOW_PWR_IDLE_S0 flag must be set. So including it unconditionally doesn't cause any problems. The AMD Modern Standby BIOS Implementation Guide defines two UUIDs, one for getting the device constraints, and one for handling notifications. This differs from the Intel specification and the linux driver implementation. For this reason I haven't implemented any of the notification callbacks yet. BUG=b:178728116 TEST=Boot OS and verify _DSM is called: [ 0.226701] lps0_device_attach: ACPI: \_SB_.PEP_: _DSM function mask: 0x3 [ 0.226722] lpi_device_get_constraints_amd: ACPI: \_SB_.PEP_: _DSM function 1 eval successful [ 0.226723] lpi_device_get_constraints_amd: ACPI: \_SB_.PEP_: LPI: constraints list begin: [ 0.226724] lpi_device_get_constraints_amd: ACPI: \_SB_.PEP_: LPI: constraints list end Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I2deef47eabe702efe1a0f3747c9f27bcec37464b Reviewed-on: https://review.coreboot.org/c/coreboot/+/52445 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-19Revert "soc/amd/cezanne: Add support to perform early EC sync"Karthikeyan Ramasubramanian
This reverts commit ad7c33abd21dfdde75c6ffa23c31cbe46826d2d5. With EFS2 already enabled in EC, enabling early EC sync is not required. Also a workaround has been added in payload to address any boot issues. BUG=b:185277224 TEST=Build and boot to OS in Guybrush in both normal and recovery mode. Cq-Depend: chromium:2832032 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Change-Id: I34f8433739754365c8e5a10fdf7e58e3d1e7e797 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52419 Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-18soc/amd/cezanne/Kconfig: add missing ACPI_BERT and ACPI_BERT_SIZEFelix Held
ACPI_BERT_SIZE is used in the FSP driver and the fsp_m_params.c. The latter one is planned to be deprecated though. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Id1a250defbd31e255df9b7a7dd8488dc3182649b Reviewed-on: https://review.coreboot.org/c/coreboot/+/52385 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-16psp_verstage: move svc to platform-specific dirKangheui Won
Since there are some differences between picasso PSP svc and cezanne PSP svc, each platform should have their own svc wrapper. Moreover cezanne PSP will drop unused parameters from update_psp_bios_dir and save_uapp_data so make wrapper around it. BUG=b:182477057 BRANCH=none TEST=build psp_verstage and boot on zork Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I69f998865fc3184ea8900a431924a315c5ee9133 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52307 Reviewed-by: Martin Roth <martinroth@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-16soc/amd: move psp_verstage to amd/commonKangheui Won
psp_verstage is not specific to picasso. There might be picasso-specific code but move everything into common as a first step. While developing psp_verstage for cezanne picasso-specific code will move back to picasso directory. BUG=b:182477057 BRANCH=none TEST=build psp_verstage on zork Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: Ifb1df0d82b972f28be2ffebd476c2553cbda9810 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52306 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Martin Roth <martinroth@google.com>
2021-04-16soc/amd/common: Correct Kconfig enable for ACPI GPIO functionsMartin Roth
The Kconfig value specified Intel instead of AMD. BUG=b:184198808 TEST=Backlight enabled in the OS Signed-off-by: Martin Roth <martinroth@chromium.org> Change-Id: I9fbdf821591ec886f383c1a5ac197f8f213c4cc4 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52384 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-16soc/amd/cezanne: Update FADT to support S0i3Jason Glenesk
Set ACPI_FADT_LOW_PWR_IDLE_S0 flag in FADT. BUG=b:178728116 TEST=Dump FACP and confirm Flags bits match expected. Change-Id: I59ef762a18903135f9daa902ba8d1e40c451e96c Signed-off-by: Jason Glenesk <jason.glenesk@amd.corp-partner.google.com> Signed-off-by: Mathew King <mathewk@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52035 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Karthik Ramasubramanian <kramasub@google.com> Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-16soc/amd/cezanne: Add modern standby option to chip configMathew King
BUG=b:178728116 Change-Id: I0d09bd4361f5f47360daf750efbc993010804902 Signed-off-by: Jason Glenesk <jason.glenesk@amd.corp-partner.google.com> Signed-off-by: Mathew King <mathewk@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52041 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Karthik Ramasubramanian <kramasub@google.com> Reviewed-by: Furquan Shaikh <furquan@google.com> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-16soc/amd/cezanne: Port ACPI p-state and c-state entries from picassoJason Glenesk
Add generate_cpu_entries to device operations. Add support to generate cpu p-state and c-state SSDT entries. BUG=b:184151560 TEST=Dump and verify SSDT entry for CPU p-states and c-states. Change-Id: I77d8078b94fb661dc045b4184955c8cbec373d12 Signed-off-by: Jason Glenesk <jason.glenesk@amd.corp-partner.google.com> Signed-off-by: Mathew King <mathewk@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52036 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-16soc/amd/cezanne: Add uart controllers to chipset.cbIvy Jian
Add uart controller to chipset.cb and leave it off by default. Turn uart0 on for console for mainboards. BUG=none TEST=builds and boot into OS Signed-off-by: Ivy Jian <ivy_jian@compal.corp-partner.google.com> Change-Id: Iaeb7fea4b92bd89331c7ae7c1c000f8d9961fe9e Reviewed-on: https://review.coreboot.org/c/coreboot/+/52287 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: EricR Lai <ericr_lai@compal.corp-partner.google.com>
2021-04-16soc/amd/cezanne: Select VBNV_CMOSRaul E Rangel
Needed so we can switch to normal mode. BUG=b:184126844 TEST=Boot guybrush in developer mode and switch to normal mode. Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I26ad160a2372484e9753a727f2b454a31e3537a0 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52224 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-15soc/amd/common/block/gpio_banks: Use configure_scimap()Kyösti Mälkki
There is no need to stash the SCI trigger register configuration and apply it at the end. Remove this to make SCI and SMI programming more symmetrical and to use available configure_scimap function instead of implementing it again, but without the additional checks. Using this function also allows removing soc_route_sci. Change-Id: Ie23da79546858282910db65182a6315ade506279 Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/43012 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-14soc/amd/stoneyridge: use common pm_set_power_failure_state functionalityFelix Held
The functionality to restore the previous power state after power was lost that could previously be enabled by selecting MAINBOARD_POWER_RESTORE in the mainboard's Kconfig can now be achieved by selecting POWER_STATE_PREVIOUS_AFTER_FAILURE in the mainboard's Kconfig instead. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I49c4a44ca2c4fa937a823c4eddf1618739c15114 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52303 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-14soc/amd/piasso/fch: use common pm_set_power_failure_state functionalityFelix Held
The functionality to restore the previous power state after power was lost that could previously be enabled by selecting MAINBOARD_POWER_RESTORE in the mainboard's Kconfig can now be achieved by selecting POWER_STATE_PREVIOUS_AFTER_FAILURE in the mainboard's Kconfig instead. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Iab9578ebea89651dc2389bf6ca93ca3f3507eb47 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52302 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-14soc/amd/common/block/pm: rework pm_set_power_failure_stateFelix Held
Picasso and Stoneyridge didn't do a read-modify-write operation on the lower nibble of PM_RTC_SHADOW_REG, but just wrote the upper nibble as all zeros. Since the upper nibble might be uninitialized before the lower nibble gets written, do what Picasso and Stoneyridge did here instead of what the reference code does. Also add a comment why and how this register behaves a bit weird. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I0bda2349e3ae84cba50b187cc773fd8a5b17f4e2 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52301 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-14soc/amd/common/block/pm: remove POWER_STATE_DEFAULT_ON_AFTER_FAILUREFelix Held
Not selecting POWER_STATE_DEFAULT_ON_AFTER_FAILURE brings Cezanne that is currently the only SoC using this functionality in line with Picasso where the default is that the board remains in power off mode after power was lost and later restored. Boards can change this behavior by selecting POWER_STATE_OFF_AFTER_FAILURE, POWER_STATE_ON_AFTER_FAILURE or POWER_STATE_PREVIOUS_AFTER_FAILURE. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Ic96f40e3c9867cd821e58d752f58b763930f6d0f Reviewed-on: https://review.coreboot.org/c/coreboot/+/52300 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-14soc/amd/common/block/pm: select HAVE_POWER_STATE_PREVIOUS_AFTER_FAILUREFelix Held
Without this being selected, mainboards can't select MAINBOARD_POWER_STATE_PREVIOUS to use the power state restoration code path in pmlib.c Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I753659fa753e03a66b6c6b2eb97e7ef20c71ca57 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52299 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-14Rename do_printk() to printk()Nico Huber
The indirection seems unnecessary. The macros throw features like `-Wmisleading-indentation` off, though. Default build for QEMU/Q35 is unchanged. Change-Id: Ie4eab935a367b5ad6b38225c4973d41d9f70ef10 Signed-off-by: Nico Huber <nico.h@gmx.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51887 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
2021-04-14soc/amd/picasso/fch: add missing amdblocks/gpio_banks.h headerFelix Held
The prototype of gpio_add_events() is provided by that header file. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: Ia384c9297ac1e24bf0b1bcce048012a247406f39 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52274 Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-14soc/amd/cezanne/fch: process ACPI PM/GPE and GPIO eventsFelix Held
BUG=b:184549804 Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I4ebbe9667d18a96b1a363d0353c612e214699d12 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52273 Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-14soc/amd/cezanne: save chipset state to CBMEMMartin Roth
Guybrush complains that this is missing during the boot, so add it to cezanne. I verified that the registers in gpio.c are correct. BUG=b:184549804 TEST=Build and boot Signed-off-by: Martin Roth <martinroth@chromium.org> Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I3de3764c99fe89b962db88065575463b365ddaf5 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51751 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-14soc/amd/picasso/romstage: factor out chipset state saving functionalityFelix Held
Since Cezanne needs the exact same code, move it to the common directory and add a Kconfig option to add this functionality to the build. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I04c4295071a3df7afcb4dfd5435b11fb0bf6963f Reviewed-on: https://review.coreboot.org/c/coreboot/+/52272 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Arthur Heymans <arthur@aheymans.xyz> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-11soc/amd/common/block/gpio: remove SoC type check in gpio_fill_wake_stateFelix Held
Verified that all accessed registers exist in all SoCs that use this code (Carrizo, Mullins, Stoneyridge, Picasso and Cezanne at the moment) and that the bit definitions match as well. Also at the time of writing this patch only Picasso calls gpio_fill_wake_state, so dropping the check won't change behavior. This also avoids having SoC specific code that doesn't get selected by Kconfig options in the common AMD SoC directory and also avoids having to add a check for SOC_AMD_CEZANNE to support this functionality on Cezanne in a follow-up patch. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: If770780a67776daf81744db1b635ffd402653a47 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52223 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-10soc/amd/cezanne: Set Power state after power failureKarthikeyan Ramasubramanian
Configure the power state to return to when the power is re-applied after power failure. BUG=b:183739671 TEST=Build and Boot to OS in Majolica and Guybrush. By default when the power fails the device turns on after power is re-applied. When the POWER_ON_AFTER_POWER_FAILURE is disabled, the device remains off even after the power is re-applied. Change-Id: I21c5da08c82156d6239450ef6921771da74cbaa1 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52049 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-10soc/amd/common: Handle power resume after power failureKarthikeyan Ramasubramanian
Introduce a power management library to handle the power resume after power failure. Enable HAVE_POWER_STATE_AFTER_FAILURE config when this library is enabled. BUG=b:183739671 TEST=Build Guybrush and Majolica mainboard. Change-Id: Iea4ea57d747425fe6714d40ba6e60f2447febf28 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51924 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2021-04-10soc/amd/cezanne: Add GRXS and GTXS methodEric Lai
Add GRXS and GTXS support. Move the gpio method into common place. Signed-off-by: Eric Lai <ericr_lai@compal.corp-partner.google.com> Change-Id: I8ba377179d6976cf26ed0dc521d8e4eff051dc85 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52202 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2021-04-08soc/amd: remove special GPIO_2 override soc_gpio_hookKyösti Mälkki
This override was added to have the SCI mapping configured if GPIO was used as WAKE_L pin. This however didn't set up the SCI level and trigger information, so it likely never worked as intended. Change-Id: I44661f05c8f517ece88714c625603579731d174b Signed-off-by: Kyösti Mälkki <kyosti.malkki@gmail.com> Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/43049 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-08soc/amd/common: Add PM_ESPI_INTR_CTRLRaul E Rangel
This register is used for masking/unmasking eSPI IRQs. BUG=none TEST=Build guybrush Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: Ia209539b2e0ce390e227757b16c2969b9124a845 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52142 Reviewed-by: Martin Roth <martinroth@google.com> Reviewed-by: Rob Barnes <robbarnes@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-07soc/amd/picasso/acpi: fix domain argument of acpigen_write_CSD_packageFelix Held
Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I50e88ac946b9d8797571f9e3d4b325db760e423f Reviewed-on: https://review.coreboot.org/c/coreboot/+/51932 Reviewed-by: Raul Rangel <rrangel@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-07soc/amd/cezanne: Pass DXIO and DDI Descriptors to FSPMatt Papageorge
This patch adds the functionality to write the DXIO and DDI descriptors to the UPD data structure to the SoC code and adds the mainboard_get_dxio_ddi_descriptors function to each mainboard using the Cezanne SoC that gets called to get the descriptors from the board code. Change-Id: I1cb36addcf0202cd56ce99e610a13d6d230bc981 Signed-off-by: Matt Papageorge <matthewpapa07@gmail.com> Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51948 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-07vc/amd/fsp/cezanne: update UPD headersMatt Papageorge
The UPD header files get generated as part of the FSP build process. For the initial Cezanne development we took the Picasso UPD data structures as a starting point. This patch replaces it with the first version of the Cezanne-specific UPD data structures that is present in version 12 of the internal work-in-progress FSP binary drops. The serial_port_stride UPD-M field is removed, since the information is already given by serial_port_use_mmio. The stride is 4 bytes for the MMIO UART case and 1 byte for the legacy I/O case. BUG=b:182524631 TEST=NVMe works on google/guybrush when the rest of the patch train is applied as well. Change-Id: Idca235029bf2e68d403230d55308820cab61a6c0 Signed-off-by: Matt Papageorge <matthewpapa07@gmail.com> Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51806 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-05soc/amd/cezanne: Add soc/msr.hRaul E Rangel
This is a copy of picasso. BUG=b:184151560 TEST=Compared with the cezanne PPR. Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: Ia4bc40daa971c126c2596837155312d411b91a06 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51983 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-05soc/amd/common: Add some ESPI register definitionsMartin Roth
Use definitions instead of magic numbers clean up some whitespace while I'm here. BUG=b:183207262 TEST=Build Change-Id: Ieae53b12e5303641fb3f180c47468aaa6906e9af Signed-off-by: Martin Roth <martinroth@chromium.org> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51747 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-04-05soc/amd: Make espi_clear_decodes privateRaul E Rangel
espi_setup already clears most of the controller registers. So this change consolidates the clear logic into one spot. This shouldn't result in a behavior change on Picasso. Picasso already has the eSPI decodes clear on boot, so this change is a nop. BUG=b:183524609 TEST=Boot guybrush to the OS Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: Ic57689e50febd29796d8ac8d99c81e41fee5b41c Reviewed-on: https://review.coreboot.org/c/coreboot/+/52064 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-05soc/amd: Make espi_configure_decodes privateRaul E Rangel
This is only ever called after espi_setup. 55861 - AMD System Peripheral Bus Overview also says that io ranges should be configured before enabling the BUS_MASTER bit. BUG=b:183524609 TEST=Boot guybrush to OS Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I074e487d8768a578ee889a125b9948e3aa6c7269 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52059 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-05soc/amd/common/espi: Clear DNCMD_COMPLETE on completionRaul E Rangel
Tidy up the interrupt status. This will leave SLAVE0_INT_STS = 0. BUG=b:183524609 TEST=Boot guybrush to OS Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I950cfb81521e35758c120a482670cfdb924201d2 Reviewed-on: https://review.coreboot.org/c/coreboot/+/52056 Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-05soc/amd/common/espi: Reset eSPI registers to known stateRaul E Rangel
This sets the eSPI registers to the reset values specified in the PPR. On Cezanne, the PSP modifies these registers such that the eSPI peripheral cannot send DEFER packets. This causes random bus errors. These reset values are identical to what is currently used on Zork. I didn't clear out ESPI_DECODE because it's currently being done by cb:51749. BUG=b:183524609 TEST=Boot guybrush to the OS Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: Ic3a9860747aac78121358b4499d8a38052236c0c Reviewed-on: https://review.coreboot.org/c/coreboot/+/52058 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-05soc/amd/common/espi: Add missing eSPI register definitionsRaul E Rangel
These are defined in the public Picasso PPR - 55570-B1 Rev 3.15. BUG=b:183524609 TEST=Build guybrush Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I7e601f767327e0a24a086146623af039388b2e7b Reviewed-on: https://review.coreboot.org/c/coreboot/+/52057 Reviewed-by: Furquan Shaikh <furquan@google.com> Reviewed-by: Rob Barnes <robbarnes@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-05soc/amd/common/espi: Add ESPI_ prefix to SLAVE0_INT_ENRaul E Rangel
This matches the other register definitions. BUG=b:183524609 TEST=Build guybrush Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I0ed92add633f294f92c6a0dde32851d01b10db3c Reviewed-on: https://review.coreboot.org/c/coreboot/+/52055 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-05soc/amd/cezanne: Clear eSPI ranges before configuring eSPIMartin Roth
The Cezanne PSP configures the eSPI with the assumption that it's a majolica, setting up both the serial port and the majolica EC IO decode ranges. Since guybrush is NOT a majolica, this doesn't work very well there. Clearing the decode ranges allows the guybrush platform to set the decode ranges needed for its EC. BUG=b:183524609 TEST=Set up eSPI on Guybrush Signed-off-by: Martin Roth <martinroth@chromium.org> Change-Id: I77cfb948cb9ae6d1cf001bd9e66cede8d93f50b5 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51749 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-05soc/amd/common: Add func to clear eSPI IO & memory decode rangesMartin Roth
Previously, the eSPI code would only add to existing decode ranges, and there wasn't any way to clear ranges. This clears all the ranges so the eSPI configuration can start fresh. BUG=b:183207262, b:183974365 TEST=Verify on Guybrush Signed-off-by: Martin Roth <martinroth@chromium.org> Change-Id: Ic4e67c40d34915505bdd5b431a064d2c7b6bbc70 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51748 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-04-02soc/amd/cezanne: Add support to perform early EC syncKarthikeyan Ramasubramanian
Ideally we would like to perform EC Software Sync in payload. But with the hardware requirement (EC_IN_RW) and firmware requirement (TPM command to get EC execution environment) not met yet, adding the support to perform early EC Software sync. With EFS2 enabled, this will also help cr50 to set the boot mode as NORMAL instead of NO_BOOT. BUG=None TEST=Build and Boot to OS in Guybrush. Ensure that the EC software sync is successfully complete. CBFS: Found 'ecrw.hash' @0x50400 size 0x20 in mcache @0x020171ec VB2:check_ec_hash() Hexp RW(active): 2dd8dbb78d0c626358a626037973a3d81982f88f3f38e7f759039bf84e05ccc6 VB2:check_ec_hash() Hmir: 2dd8dbb78d0c626358a626037973a3d81982f88f3f38e7f759039bf84e05ccc6 <snip> VB2:check_ec_hash() Heff RW(active): 2dd8dbb78d0c626358a626037973a3d81982f88f3f38e7f759039bf84e05ccc6 VB2:sync_ec() select_rw=RW(active) Change-Id: I820e651c6b22a833fef6f17a4ceb5a8cfb6f1616 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/52008 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-04-01soc/amd/cezanne: Enable GENERIC_GPIO_LIBRaul E Rangel
Needed so we write the correct resource into the ACPI tables. BUG=b:183737011 TEST=Boot OS and see GPIO devices working Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I2ba4349e0ed500912db40aa6ef9b649046f4358f Reviewed-on: https://review.coreboot.org/c/coreboot/+/51961 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Karthik Ramasubramanian <kramasub@google.com> Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-04-01soc/amd/cezanne: Add device tree support for I2CRaul E Rangel
This allows the cr50 on guybrush to show up in ACPI. BUG=b:183737011 TEST=Boot OS and see I2C devices initialized Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: Ifb5679b7bbefbf753217981874bb1bdaef35f6db Reviewed-on: https://review.coreboot.org/c/coreboot/+/51958 Reviewed-by: Karthik Ramasubramanian <kramasub@google.com> Reviewed-by: Mathew King <mathewk@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-31soc/amd/common/block/graphics: Don't add VBIOS to cbmem when using GOPRaul E Rangel
pci_rom_ssdt reloads the oprom from cbfs. It then places it into cbmem and writes the offsets as the ROM ACPI node. The GOP driver modifies the VBIOS so we don't want to reread from cbfs. When using GOP we also pass the offsets with the VFCT table. Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: Iaf53e750564f1f0e115cd354790da62e672d74b7 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51541 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Nikolai Vyssotski <nikolai.vyssotski@amd.corp-partner.google.com> Reviewed-by: Martin Roth <martinroth@google.com>
2021-03-31soc/amd/picasso/acpi: pass correct enum to acpigen_write_CSD_packageFelix Held
The coordtype parameter of acpigen_write_CSD_package expects a CSD_coord enum value, but HW_ALL that got passed as parameter is a PSD_coord enum value, so replace that with the correct CSD_HW_ALL enum value. TEST=Timeless build results in identical binary for Mandolin. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Found-by: Paul Menzel <pmenzel@molgen.mpg.de> Change-Id: I90b19345b8dc6d386b6acfa81c6c072dcd6981ca Reviewed-on: https://review.coreboot.org/c/coreboot/+/51931 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com>
2021-03-30soc/amd/cezanne: Comment the AOAC register accessKarthikeyan Ramasubramanian
Causing the AOAC register access as part of system suspend (S3) causes the suspend procedure to be stuck. Comment it for now to unblock entering S3 and collecting the power numbers. BUG=b:181766974 TEST=Build and boot to OS in Majolica. Enter S3 through "echo mem > /sys/power/state". Change-Id: Ie93bbe393b209b784b9a2257f3916b29d84b25d1 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51926 Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Mathew King <mathewk@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-29soc/amd: smbus: Use correct type for uintptr_tPaul Menzel
Fix the format warning below by using `PRIxPTR`, which is defined as unsigned long. src/soc/amd/common/block/smbus/smbus.c:33:56: error: format specifies type 'size_t' (aka 'unsigned int') but the argument has type 'uintptr_t' (aka 'unsigned long') [-Werror,-Wformat] printk(BIOS_ERR, "Invalid SMBus or ASF base %#zx\n", mmio); ~~~~ ^~~~ %#lx src/include/console/console.h:60:61: note: expanded from macro 'printk' #define printk(LEVEL, fmt, args...) do_printk(LEVEL, fmt, ##args) ~~~ ^~~~ 1 error generated. Change-Id: I727c490d3097dcf36cdbcd4db2852cd49d11785f Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51843 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Felix Held <felix-coreboot@felixheld.de>
2021-03-29soc/amd/cezanne: factor out UPD-M configuration from romstageFelix Held
Move the parts of romstage.c that populate the UPD-M data structure to the newly created fsp_m_params.c file. Since platform_fsp_memory_init_params_cb gets called from the FSP driver and not directly from car_stage_entry the two code parts in romstage.c weren't directly interacting. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I1f7f5879ac318372042ff703ebbe584ce1c32c91 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51835 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-03-29soc/amd/picasso: factor out UPD-M configuration from romstageFelix Held
Move the parts of romstage.c that populate the UPD-M data structure to the newly created fsp_m_params.c file. Since platform_fsp_memory_init_params_cb gets called from the FSP driver and not directly from car_stage_entry the two code parts in romstage.c weren't directly interacting. Since soc/romstage.h only contains the mainboard_updm_update function prototype, rename it to soc/fsp.h. This patch also removes a few unused includes. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I52c21f13520dbdfab37587d17b3a8a3b1a780f36 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51834 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-03-29soc/amd/cezanne,picasso: rename fsp_params.c to fsp_s_params.cFelix Held
This file populates the UPD-S data structure that gets passed to the FSP-S, so add that s part to make it a bit clearer which FSP parameters it'll set up. This is also a preparation to add a fsp_m_params.c file in the following patches. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I53786df0909055e66eac675b5580909b7960944f Reviewed-on: https://review.coreboot.org/c/coreboot/+/51833 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org>
2021-03-29soc/amd: add DISABLE_KEYBOARD_RESET_PIN optionFelix Held
The KBRST_L pin will cause a reset when driven or pulled low even when the GPIO mux is set to GPIO and not native function. So when you want to use that pin as general purpose output the keyboard reset input functionality needs to be disabled by selecting this option in the board's Kconfig file to avoid causing a reset by writing a 0 to the output level bit when it's configured as an output. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Suggested-by: Furquan Shaikh <furquan@google.com> Change-Id: I517ad551db9321f26afdba15d97ddb61be1f7d51 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51757 Reviewed-by: Furquan Shaikh <furquan@google.com> Reviewed-by: Kangheui Won <khwon@chromium.org> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-29soc/amd/cezanne: Implement PROVIDES_ROM_SHARINGRaul E Rangel
BUG=none TEST=Build guybrush and verified with the PPR that the register and bits are still the same Signed-off-by: Raul E Rangel <rrangel@chromium.org> Change-Id: I0619f84cf82cbb90ded9dfd58afa6acc9520fb8e Reviewed-on: https://review.coreboot.org/c/coreboot/+/51780 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-03-29soc/amd/common/block/acpimmio/mmio_util: add fch_disable_kb_rstFelix Held
Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Suggested-by: Kangheui Won <khwon@chromium.org> Change-Id: Ie65e39ffb8c353415f5b68e1e0f378d18eeb7498 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51784 Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-29soc/amd: move PM_RST_CTRL1 register definition to common acpimmio headerFelix Held
TEST=Verified that this register and the defined bits exist in Cezanne, Picasso, Stoneyridge, Bolton and SB800. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I32d1d577b05edab006981516a5aefd822e7b984a Reviewed-on: https://review.coreboot.org/c/coreboot/+/51783 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-03-29soc/amd/*/gpio: include types.h instead of stdint.h to have size_tFelix Held
Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: I7a747d4c28e6d449c054ce83966767e13b51a939 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51782 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-03-28soc/amd/common/gpio: add PAD_NF_SCI pad typeFelix Held
This patch adds a pin configuration macro that supports both switching a pin to its native function and configuring it as a SCI source. This is a preparation to remove the GPIO2 soc_gpio_hook. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Change-Id: If0da5c010f35fd902f6b8857368daec93c12394a Reviewed-on: https://review.coreboot.org/c/coreboot/+/50373 Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-28soc/amd/picasso/mca: add missing comma in mca_bank_name array of stringsFelix Held
Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Found-by: Coverity CID 1451389 Change-Id: I0af379360fc95e4c6b72d677738c6e7497ed9206 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51788 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com>
2021-03-28soc/amd/common/block/i2c: Use `size_t` for `num_pins`Angel Pons
There's no need to use a fixed-width type here. Change-Id: I727c64661990040db356c5508fecc0a65960c095 Signed-off-by: Angel Pons <th3fanbus@gmail.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51794 Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-28soc/amd/common/block/i2c: Fix printf format specifiersAngel Pons
The correct printf format specifier for an `unsigned int` is `%u`. Change-Id: Iaf780eb366f8c3493b89beb9a5643fa285e7825d Signed-off-by: Angel Pons <th3fanbus@gmail.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51793 Reviewed-by: Karthik Ramasubramanian <kramasub@google.com> Reviewed-by: Marshall Dawson <marshalldawson3rd@gmail.com> Reviewed-by: Paul Menzel <paulepanter@users.sourceforge.net> Reviewed-by: Felix Held <felix-coreboot@felixheld.de> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-25soc/amd/common/block/i2c: fix control flow bugFelix Held
commit 4f87ae1d4a3a597f1260534001bd99160cc8ca99 introduced a regression in the I2C initialization resulting in soc_i2c_misc_init never getting called, since the continue statement was indented like it belonged to the if above, but due to the missing curly braces it was outside the if block. Signed-off-by: Felix Held <felix-coreboot@felixheld.de> Found-by: Coverity CID 1451395, 1451387 Change-Id: Id1f17ad59cba44e96881f5511df303ae90841ab3 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51786 Reviewed-by: Karthik Ramasubramanian <kramasub@google.com> Reviewed-by: Raul Rangel <rrangel@chromium.org> Reviewed-by: Angel Pons <th3fanbus@gmail.com> Reviewed-by: Furquan Shaikh <furquan@google.com> Tested-by: build bot (Jenkins) <no-reply@coreboot.org>
2021-03-25soc/amd/common/block/gpio_defs: Wake from either S0i3 or S3Karthikeyan Ramasubramanian
Add a helper bit mask to enable wake from either S0i3 or S3. BUG=None TEST=Build the Guyrbush mainboard. Change-Id: I934abad78135260081a61aee4c496b362e483de1 Signed-off-by: Karthikeyan Ramasubramanian <kramasub@google.com> Reviewed-on: https://review.coreboot.org/c/coreboot/+/51745 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Furquan Shaikh <furquan@google.com>
2021-03-24mb/google/guybrush: disable KBRSTENKangheui Won
GPIO129 is muxed with KBRST, so setting GPIO129 to low causes reset when KBRSTEN is set to 1. Since reset value of KBRSTEN is 1 we need a logic to clear it. BUG=b:183340503 TEST=build Signed-off-by: Kangheui Won <khwon@chromium.org> Change-Id: I194e8432a14d6105f6bcf12111647f5aad4e2de2 Reviewed-on: https://review.coreboot.org/c/coreboot/+/51727 Tested-by: build bot (Jenkins) <no-reply@coreboot.org> Reviewed-by: Martin Roth <martinroth@google.com> Reviewed-by: Raul Rangel <rrangel@chromium.org>