Matt DeVillier has submitted this change. ( https://review.coreboot.org/c/coreboot/+/86026?usp=email )
(
5 is the latest approved patch-set.
No files were changed between the latest approved patch-set and the submitted one.
)Change subject: mb/google/volteer/variants/drobit: fix power_limits_config
......................................................................
mb/google/volteer/variants/drobit: fix power_limits_config
Drobit shows little power usage (around 9W) and very low clock speeds
(1.1GHz - 1.4GHz) under load (`stress -c $(nproc)`), despite being at
temperature too low for thermal throttling (40-50C). It turns out that
power_limits_config is set to the lower end of the dptf power limit
ranges as opposed to baseboard and other variants. This seems to
prevent the device from using the intended power limits.
Tested: Boot and confirm more reasoable power usage (17W) and clock
speeds (around 2.5GHz) as well as good temperatures (topped 85C) and
stability under 100% load (`stress -c $(nproc)` for 30min).
Device for tests is i5-1135G7 and 16GG RAM.
Change-Id: Id0478c713b51db4972e7d93ec597a30fa885c22b
Signed-off-by: Ingo Reitz <9l(a)9lo.re>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/86026
Reviewed-by: Matt DeVillier <matt.devillier(a)gmail.com>
Tested-by: build bot (Jenkins) <no-reply(a)coreboot.org>
---
M src/mainboard/google/volteer/variants/drobit/overridetree.cb
1 file changed, 2 insertions(+), 2 deletions(-)
Approvals:
build bot (Jenkins): Verified
Matt DeVillier: Looks good to me, approved
diff --git a/src/mainboard/google/volteer/variants/drobit/overridetree.cb b/src/mainboard/google/volteer/variants/drobit/overridetree.cb
index c791079..148af6e 100644
--- a/src/mainboard/google/volteer/variants/drobit/overridetree.cb
+++ b/src/mainboard/google/volteer/variants/drobit/overridetree.cb
@@ -12,8 +12,8 @@
register "tcc_offset" = "8"
register "power_limits_config[POWER_LIMITS_U_4_CORE]" = "{
- .tdp_pl1_override = 9,
- .tdp_pl2_override = 28,
+ .tdp_pl1_override = 17,
+ .tdp_pl2_override = 64,
.tdp_pl4 = 105,
}"
--
To view, visit https://review.coreboot.org/c/coreboot/+/86026?usp=email
To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: coreboot
Gerrit-Branch: main
Gerrit-Change-Id: Id0478c713b51db4972e7d93ec597a30fa885c22b
Gerrit-Change-Number: 86026
Gerrit-PatchSet: 7
Gerrit-Owner: Ingo Reitz <9l(a)9lo.re>
Gerrit-Reviewer: Ariel Chang <ariel_chang(a)pegatron.corp-partner.google.com>
Gerrit-Reviewer: Matt DeVillier <matt.devillier(a)gmail.com>
Gerrit-Reviewer: Nick Vaccaro <nvaccaro(a)chromium.org>
Gerrit-Reviewer: Paul Yang <paul.f.yang(a)intel.corp-partner.google.com>
Gerrit-Reviewer: Wayne3 Wang <wayne3_wang(a)pegatron.corp-partner.google.com>
Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org>
Gerrit-CC: Paul Menzel <paulepanter(a)mailbox.org>
Matt DeVillier has submitted this change. ( https://review.coreboot.org/c/coreboot/+/85809?usp=email )
Change subject: soc/intel/xeon_sp/ebg/soc_xhci: Check if BAR is reachable
......................................................................
soc/intel/xeon_sp/ebg/soc_xhci: Check if BAR is reachable
On x86_32 the xHCI BAR isn't reachable as it's mapped in high MMIO.
Currently this is not a problem since the code is unused.
Add a check and return NULL instead of cutting of the higher bits
and thus do not return an invalid pointer. On x86_64 it's working
when the extended page-tables are installed.
Change-Id: I00496ad476c33e0984d7cb0019f27154302edda5
Signed-off-by: Patrick Rudolph <patrick.rudolph(a)9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/85809
Tested-by: build bot (Jenkins) <no-reply(a)coreboot.org>
Reviewed-by: Shuo Liu <shuo.liu(a)intel.com>
---
M src/soc/intel/xeon_sp/ebg/soc_xhci.c
1 file changed, 7 insertions(+), 0 deletions(-)
Approvals:
build bot (Jenkins): Verified
Shuo Liu: Looks good to me, approved
diff --git a/src/soc/intel/xeon_sp/ebg/soc_xhci.c b/src/soc/intel/xeon_sp/ebg/soc_xhci.c
index f8aa37b..f7b1320 100644
--- a/src/soc/intel/xeon_sp/ebg/soc_xhci.c
+++ b/src/soc/intel/xeon_sp/ebg/soc_xhci.c
@@ -17,6 +17,13 @@
printk(BIOS_ERR, "XHCI BAR is not found\n");
return NULL;
}
+
+#if ENV_X86_32
+ assert(res->base < 0x100000000ULL);
+ if (res->base >= 0x100000000ULL)
+ return NULL;
+#endif
+
return (void *)(uintptr_t)res->base;
}
--
To view, visit https://review.coreboot.org/c/coreboot/+/85809?usp=email
To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: coreboot
Gerrit-Branch: main
Gerrit-Change-Id: I00496ad476c33e0984d7cb0019f27154302edda5
Gerrit-Change-Number: 85809
Gerrit-PatchSet: 4
Gerrit-Owner: Patrick Rudolph <patrick.rudolph(a)9elements.com>
Gerrit-Reviewer: Christian Walter <christian.walter(a)9elements.com>
Gerrit-Reviewer: Johnny Lin <Johnny_Lin(a)wiwynn.com>
Gerrit-Reviewer: Jonathan Zhang <jon.zhixiong.zhang(a)gmail.com>
Gerrit-Reviewer: Matt DeVillier <matt.devillier(a)gmail.com>
Gerrit-Reviewer: Shuo Liu <shuo.liu(a)intel.com>
Gerrit-Reviewer: Tim Chu <Tim.Chu(a)quantatw.com>
Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org>
Matt DeVillier has submitted this change. ( https://review.coreboot.org/c/coreboot/+/85680?usp=email )
Change subject: Documentation: Add Topton N100 (X2F)
......................................................................
Documentation: Add Topton N100 (X2F)
Document the board and how to flash coreboot.
Change-Id: Id585b064054b338ea8cead6edb6c5153030b9cde
Signed-off-by: Alicja Michalska <alicja.michalska(a)9elements.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/85680
Tested-by: build bot (Jenkins) <no-reply(a)coreboot.org>
Reviewed-by: Nicholas Chin <nic.c3.14(a)gmail.com>
---
M Documentation/mainboard/index.md
A Documentation/mainboard/topton/adl/x2f-n100.md
2 files changed, 108 insertions(+), 0 deletions(-)
Approvals:
build bot (Jenkins): Verified
Nicholas Chin: Looks good to me, approved
diff --git a/Documentation/mainboard/index.md b/Documentation/mainboard/index.md
index 4cbcaf6..f9f9048 100644
--- a/Documentation/mainboard/index.md
+++ b/Documentation/mainboard/index.md
@@ -408,6 +408,13 @@
Beaglebone Black <ti/beaglebone-black.md>
```
+## Topton
+```{toctree}
+:maxdepth: 1
+
+X2F-N100 <topton/adl/x2f-n100.md>
+```
+
## UP
```{toctree}
diff --git a/Documentation/mainboard/topton/adl/x2f-n100.md b/Documentation/mainboard/topton/adl/x2f-n100.md
new file mode 100644
index 0000000..c386ad3
--- /dev/null
+++ b/Documentation/mainboard/topton/adl/x2f-n100.md
@@ -0,0 +1,101 @@
+# Topton N100 (X2F)2 Firewall Appliance
+
+This page describes how to run coreboot on the Topton N100.
+
+```{eval-rst}
++---------------+---------------------------------+----------------------+
+| Binary file | Apply | Required / Optional |
++===============+=================================+======================+
+| IFD | Intel Flash Descriptor | Required (see below) |
++---------------+---------------------------------+----------------------+
+| ME | Intel Management Engine | Required (see below) |
++---------------+---------------------------------+----------------------+
+| FSP-M & FSP-S | Intel Firmware Support Package | Required |
++---------------+---------------------------------+----------------------+
+```
+
+## Flashing coreboot
+
+**WARNING:** There are multiple devices from the same vendor with
+similar name, using different board layout, capabilities and EC/SuperIO.
+
+Likewise, there are other vendors selling their devices with
+the same mainboard. According to the sticker on the box: `H30W-N100-226`.
+
+Please **DO NOT cross-flash firmware on "similar-looking" device**.
+Doing so **may kill your device**. You have been warned :)
+
+### Internally
+
+Vendor of this motherboard hasn't locked any flash regions, resulting
+in [flashprog] having full access to the SPI chip.
+Assuming that user had booted Linux with `iomem=relaxed`, they can:
+- Flash coreboot from stock firmware
+- Flash stock firmware from coreboot
+- Update coreboot build to a newer version
+
+ Without opening the case and connecting the SPI flasher.
+
+Please note that for AlderLake-N platform you will need to use
+[flashprog] v1.3.0 or newer.
+
+[flashrom] is broken due to regressions, which results in
+failed flashes, bricking the device.
+
+[flashprog] is a better maintained fork of [flashrom], which
+works flawlessly.
+
+You can skip extracting `SI_BIOS` and `SI_ME` regions from your ROM
+and flash coreboot to `SI_BIOS` region by issuing the following command:
+`flashprog -p internal --ifd -i SI_BIOS -w ./build/coreboot.rom`
+
+### Externally
+
+SPI chip on this mainboard is located near the SoC, on the other side
+of the board (upper-right corner).
+Please note that SPI voltage on this board is standard 3.3V, despite
+using mobile SoC.
+Vendor populated this board with Winbond W25Q128JV chip in SOIC-8 package.
+
+## Functionality
+
+### Tested and working
+
+- All USB ports (including mPCIe)
+- All NICs (4x Intel I226-V 2.5GbE)
+- All display outputs (HDMI, DisplayPort)
+- M.2 NVME (PCIe x2 electrically)
+- mPCIe WiFi
+- SATA port
+- Intel PTT (fTPM 2.0)
+- PC Speaker (goes beep-boop)
+- Serial console on the front of the device (RS232 in RJ45 form-factor),
+"Cisco-style"
+- PCIe passtrough (NICs to VMs, such as OpenWrt using libvirt)
+
+- Payload: EDK2, LinuxBoot
+- OS: Alpine Linux, Windows 11
+
+### Untested, broken
+
+- 5G modem on special M.2-like connector (lack hardware for it).
+- Suspend in Windows 11
+(might work, but Windows has been tested from USB drive).
+
+## Specification
+```{eval-rst}
++--------+-------------------------------+
+| SoC | Intel AlderLake N100 (IoT) |
++--------+-------------------------------+
+| Memory | DDR5 SO-DIMM (single-channel) |
++--------+-------------------------------+
+| EC | ITE IT8625E |
++--------+-------------------------------+
+| SPI | Winbond W25Q128 (16MiB 3.3V) |
++--------+-------------------------------+
+| NIC | 4x Intel I226-V (2.5GbE) |
++--------+-------------------------------+
+```
+
+[flashprog]: https://flashprog.org/wiki/Flashprog
+[flashrom]: https://flashrom.org/Flashrom
--
To view, visit https://review.coreboot.org/c/coreboot/+/85680?usp=email
To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: coreboot
Gerrit-Branch: main
Gerrit-Change-Id: Id585b064054b338ea8cead6edb6c5153030b9cde
Gerrit-Change-Number: 85680
Gerrit-PatchSet: 4
Gerrit-Owner: Alicja Michalska <ahplka19(a)gmail.com>
Gerrit-Reviewer: Matt DeVillier <matt.devillier(a)gmail.com>
Gerrit-Reviewer: Nicholas Chin <nic.c3.14(a)gmail.com>
Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org>
Matt DeVillier has submitted this change. ( https://review.coreboot.org/c/coreboot/+/81611?usp=email )
Change subject: Documentation: Add Erying Polestar G613 Pro
......................................................................
Documentation: Add Erying Polestar G613 Pro
Document the board and process of building/flashing coreboot on it.
Change-Id: I5d60508dbde10373b0da2fb4ece0992760d3121c
Signed-off-by: Alicja Michalska <ahplka19(a)gmail.com>
Reviewed-on: https://review.coreboot.org/c/coreboot/+/81611
Reviewed-by: Nicholas Chin <nic.c3.14(a)gmail.com>
Tested-by: build bot (Jenkins) <no-reply(a)coreboot.org>
---
A Documentation/mainboard/erying/tgl/tgl_matx.md
M Documentation/mainboard/index.md
2 files changed, 211 insertions(+), 0 deletions(-)
Approvals:
build bot (Jenkins): Verified
Nicholas Chin: Looks good to me, approved
diff --git a/Documentation/mainboard/erying/tgl/tgl_matx.md b/Documentation/mainboard/erying/tgl/tgl_matx.md
new file mode 100644
index 0000000..1234a1c
--- /dev/null
+++ b/Documentation/mainboard/erying/tgl/tgl_matx.md
@@ -0,0 +1,203 @@
+# Erying Polestar G613 Pro
+
+This page describes how to run coreboot on the [Erying Polestar G613 Pro].
+
+## Required proprietary blobs
+
+To build full working image of coreboot, the following blobs are required:
+
+```{eval-rst}
++-----------------+---------------------------------+----------------------+
+| Binary file | Apply | Required / Optional |
++=================+=================================+======================+
+| Microcode | CPU Microcode | Required (see notes) |
++-----------------+---------------------------------+----------------------+
+| FSP-M & FSP-S | Intel Firmware Support Package | Required |
++-----------------+---------------------------------+----------------------+
+| ME | Intel Management Engine | Required (see notes) |
++-----------------+---------------------------------+----------------------+
+| FD | Intel Flash Descriptor | Required (see notes) |
++-----------------+---------------------------------+----------------------+
+```
+
+Microcode for those SoCs cannot be generated from the tree.
+While boards with D1 (production) stepping may work, microcode Intel had
+included in their tree is too old, which causes issues with APIC
+(Advanced Programmable Interrupt Controller), resulting in overall instability.
+
+This is **required** for boards sold with D0 SoC revision (Engineering Sample).
+Maintainer of this port had included publicly-available [microcodes] in
+`3rdparty/blobs` coreboot repository, which are being pulled as submodule.
+
+To choose appropriate microcode for your system, you should choose:
+1. If your motherboard uses Engineering Sample (D0) stepping:
+ - `cpu806D0_platC2_ver00000054_2021-05-07_PRD_B0F9E245.bin`
+2. If your motherboard uses retail (D1) stepping:
+ - `cpu806D1_platC2_ver00000046_2023-02-27_PRD_08E6188A.bin`
+
+By going to `Chipset -> Include CPU microcode in CBFS
+(Include external microcode binary)`
+
+Failure to choose an appropriate microcode may result in:
+- Bricked (unbootable) board
+- Issues with APIC, resulting in random freezes
+- MCE (Machine Check Exception) errors
+- Unstable system RAM, leading to bit flips and data corruption
+
+There are no extra steps required for FSP.
+Both SKUs work perfectly with FSP Intel publishes in their public repository.
+coreboot automatically pulls FSP binary from `3rdparty/fsp` submodule
+at build time, which points to TigerLakeFspBinPkg/Client package in
+official Intel FSP repository.
+
+## Flashing coreboot
+
+### Internally
+
+Vendor of this motherboard hasn't locked any flash regions, resulting in
+[flashrom] having full access to the SPI chip.
+Assuming that user had booted Linux with `iomem=relaxed`, they can:
+- Flash coreboot from stock firmware
+- Flash stock firmware from coreboot
+- Update coreboot build to a newer version
+
+Without opening the case and connecting the SPI flasher.
+Please note that you will need to use [flashrom] v1.3.0 or newer,
+as older versions won't detect the chipset.
+
+If you're using [flashrom] or [flashprog] (fork of flashrom), you can
+skip extracting `SI_BIOS` and `SI_ME` regions from your ROM, and flash
+coreboot to `SI_BIOS` region by issuing the following command:
+`flashrom -p internal --ifd -i SI_BIOS -w ./build/coreboot.rom`
+
+### Externally
+
+SPI chip on this mainboard is located right underneath the PCH heatsink.
+Due to it's location, it's likely that you will need to move (or remove)
+the PCH (Platform Controller Hub) heatsink in order to clip it properly.
+
+Vendor populated this board with Winbond W25Q128FV chip in SOIC-8 package.
+Although the stock chip is 16MiB, it is possible to replace it with 32MiB
+equivalent if user desires to use LinuxBoot payload instead of EDK2.
+
+Please note that SPI voltage on this board is standard 3.3V,
+despite using mobile SoC and PCH.
+
+## Tested and working
+
+- RS232 serial output from IT8613E for debugging (cbmem, Linux)
+- Fan control from userspace (IT8613E Environment Controller)
+- USB2.0 and 3.0
+- HDMI (iGPU, including audio)
+- Realtek RTL8111 (GbE NIC)
+- Realtek ALC897 (integrated audio)
+- PCIe x16 4.0 (SoC)
+- PCIe x1 3.0 (PCH)
+- M.2 x4 4.0 (SoC)
+- M.2 x4 3.0 (PCH)
+- M.2 NGFF (WiFi)
+- 4x SATA3 ports
+- IOMMU/VT-x (PCIe passthrough)
+- PCIe ReBAR (Resizable BAR)
+- Intel PTT (fTPM 2.0)
+
+## Work in progress, broken, issues
+
+- XMP Profiles (some people reported issues, despite successful tests).
+You can enable it by setting `SpdProfileSelected` in `romstage_fsp_params.c`.
+See [FSP XMP flags] for configuration options, proceed with caution.
+- GOP init on external GPUs (most EDK2 branches do not include module
+necessary to load external Option ROMs)
+- Sleep states (which were broken on stock as well)
+- USB3.2 might take few tries to get detected at full speed
+- iGPU DisplayPort (very simple fix, did not have time to fix GPIO)
+- Automatic fan control (fans will always spin at 50% - see below)
+- 2x USB2.0 FP and M.2 NGFF USB2.0 not mapped (yet)
+- PCIe ASPM (results in AER spam in dmesg)
+
+Please ensure to:
+- Disable sleep state in your OS to prevent data loss
+- Configure automatic fan control using pwmconfig
+(`modprobe it87 force_id=0x8603`)
+- Append `pcie_aspm=off` to your kernel commandline to avoid dmesg spam.
+
+## Notes
+
+1. Required blobs, if flashing the entire flash chip.
+ They can be skipped safely if you are planning on flashing
+ only the `SI_BIOS` region.
+ - Intel Flash Descriptor (IFD): `descriptor.bin`
+ - Intel Management Engine (ME): `me.bin`
+
+ Both blobs included in `3rdparty/blobs` repository were extracted
+ from vendor's firmware (available to download on their website).
+ IFD region has been modified using `ifdtool` to set
+ `MeAltDisable` flag.
+
+2. Modifications
+
+ It is possible to replace Winbond 16MB chip with 32MB equivalent,
+ which would allow you to use LinuxBoot or implement
+ RO + A/B VBOOT update scheme.
+
+3. Payload and pre-OS display output
+
+ If you are using an external graphics card (AMD, Nvidia, Intel Arc),
+ you will see output in your OS as soon as kernel initializes the
+ card (called "modprobing" in Linux) regardless of payload you chose.
+
+ This board was tested with following payloads:
+ - EDK2
+ - U-Boot
+ - LinuxBoot (U-Root + Linux kernel)
+
+If you would like to see output on your iGPU before that stage
+(for picking a different boot medium or toggling Secure Boot setting),
+you need to use [MrChromebox's EDK2] fork and include [GOP driver] for
+TigerLake iGPU in your build.
+
+This will allow you to see output of EDK2 (payload, boot picker)
+on your monitor connected to iGPU.
+
+If you're planning to primarly use an external card, disable iGPU by
+enabling `Chipset -> Disable Integrated GFX Controller (0:2:0)`
+and use [elly's EDK2] tree.
+
+In order to enable loading Option ROMs from PCIe devices, go to:
+`Payload -> edk2 additional custom build parameters`
+and add the string: `-D LOAD_OPTION_ROMS=TRUE`
+
+This functionality has been tested with following graphics cards,
+with following results:
+- Nvidia GeForce RTX3080, RTX3090: Works perfectly
+- AMD Radeon RX6600XT, RX7800XT: Works with ReBAR disabled,
+no output in EDK2 with ReBAR enabled
+- Intel Arc A580: Works with ReBAR disabled,
+corrupted framebuffer before modprobing with ReBAR enabled
+
+## Specification
+
+```{eval-rst}
++------------------+------------------------------+
+| SoC | Intel TigerLake Halo |
++------------------+------------------------------+
+| PCH | Intel HM570 |
++------------------+------------------------------+
+| Super I/O | ITE IT8613E |
++------------------+------------------------------+
+| SPI | Winbond W25Q128FV 16MiB 3.3V |
++------------------+------------------------------+
+| NIC | Realtek RTL8111 |
++------------------+------------------------------+
+| Audio | Realtek ALC897 |
++------------------+------------------------------+
+```
+
+[Erying Polestar G613 Pro]: https://www.erying.cc/products-detail/id-97.html
+[microcodes]: https://github.com/platomav/CPUMicrocodes/tree/master/Intel
+[FSP XMP Flags]: https://github.com/intel/FSP/blob/master/TigerLakeFspBinPkg/Client/Include/…
+[MrChromebox's EDK2]: https://github.com/MrChromebox/edk2
+[elly's EDK2]: https://github.com/ellyq/edk2
+[GOP driver]: https://github.com/MrChromebox/blobs/blob/master/soc/intel/tgl/IntelGopDriv…
+[flashrom]: https://flashrom.org/Flashrom
+[flashprog]: https://flashprog.org/wiki/Flashprog
diff --git a/Documentation/mainboard/index.md b/Documentation/mainboard/index.md
index 0009ef4..4cbcaf6 100644
--- a/Documentation/mainboard/index.md
+++ b/Documentation/mainboard/index.md
@@ -97,6 +97,14 @@
QEMU POWER9 <emulation/qemu-power9.md>
```
+## Erying
+
+```{toctree}
+:maxdepth: 1
+
+Polestar G613 Pro <erying/tgl/tgl_matx.md>
+```
+
## Facebook
```{toctree}
--
To view, visit https://review.coreboot.org/c/coreboot/+/81611?usp=email
To unsubscribe, or for help writing mail filters, visit https://review.coreboot.org/settings?usp=email
Gerrit-MessageType: merged
Gerrit-Project: coreboot
Gerrit-Branch: main
Gerrit-Change-Id: I5d60508dbde10373b0da2fb4ece0992760d3121c
Gerrit-Change-Number: 81611
Gerrit-PatchSet: 9
Gerrit-Owner: Alicja Michalska <ahplka19(a)gmail.com>
Gerrit-Reviewer: Matt DeVillier <matt.devillier(a)gmail.com>
Gerrit-Reviewer: Nicholas Chin <nic.c3.14(a)gmail.com>
Gerrit-Reviewer: build bot (Jenkins) <no-reply(a)coreboot.org>
Gerrit-CC: Angel Pons <th3fanbus(a)gmail.com>
Gerrit-CC: Felix Singer <service+coreboot-gerrit(a)felixsinger.de>
Gerrit-CC: coreboot org <coreboot.org(a)gmail.com>