coreboot-kgpe-d16/util/autoport/readme.md

16 KiB

Porting coreboot using autoport

Supported platforms

Chipset

For any sandybridge or ivybridge platform generated result should be bootable, possibly with minor fixes.

EC

EC support is likely to work on intel-based thinkpads. Other laptops are likely to miss EC support

How to use

  • Go into BIOS setup on target machine and enable all the devices. This will allow autoport to detect as much as possible

  • Boot into target machine under GNU/Linux

  • Make sure that following components are installed:

    • GCC
    • golang
    • lspci
    • dmidecode
    • acpidump
  • Grab coreboot tree

  • Execute following commands starting from coreboot tree

      cd util/ectool
      make
      cd ../inteltool
      make
      cd ../autoport
      go build
      sudo ./autoport --input_log=logs --make_logs --coreboot_dir=../..
    

    Note: in case you have problems getting gcc and golang to target machine you can just compile on another machine and transfer binaries autoport, inteltool and ectool. You'll still need other prerequisites but you may place them in the same directory as autoport.

  • Look for output unknown PCI devices. E.g.

      Unknown PCI device 8086:0085, assuming removable
    

If autoport says assuming removable then you're fine. If it doesn't then you may want to add relevant PCIIDs to autoport. When rerunning you can skip argument --make_logs to reuse the same logs

  • At this point the new board is added to the tree but don't flash it yet as it will brick you machine. Instead keep this new port and logs from util/autoport/logs somewhere safe

  • Disassemble your laptop and locate flash chip http://flashrom.org/Technology is a great resource. Flash chip is usually in SOIC-8 (2x4 pins) or SOIC-16 (2x8 chips). You'll probably have several candidates. Look up what's written on them and look up what's this chip on the web.

  • Once you know what's the chip is, get external flasher and read it. Twice. Compare the results and retry if they differ. Save the result somewhere safe, in preference copy to read-only storage as backup.

  • Compile coreboot with console enabled (EHCI debug or serial if present are recommended)

  • For recent intel chipsets you need to avoid overwriting ME firmware. Recommended procedure is (replace 8 with your flash size in MiB):

      cp backup.rom flash.rom
      dd if=coreboot/build/coreboot.rom skip=$[8-1] seek=$[8-1] bs=1M of=flash.rom
    
  • Flash the result

  • Boot and grab the log and fix the issues. See next section for useful info.

  • grep your board for FIXME. autoport adds comments when it's unsure. Sometimes it's just a minor check and sometimes it needs more involvment. See next section.

  • Send new board to review.coreboot.org. I mean it, your effort is very appreciated.

Manual fixes

SPD

If you're able to use full memory with any combination of inserted modules than this is most likely correct. In order to initialize the memory coreboot needs to know RAM timings. For socketed RAM it's stored in a small EEPROM chip which can be accessed through SPD. Unfortunately mapping between SPD addresses and RAM slots differs and cannot always be detected automatically. Resulting SPD map is encoded in function mainboard_get_spd in early_southbridge.c. autoport uses the most common map 0x50, 0x51, 0x52, 0x53 except for lenovos which are known to use 0x50, 0x52, 0x51, 0x53. To detect the correct memory map the easiest way is with vendor BIOS to boot with just one module in channel 0 slot 0 and then see where does it show up in SPD. Under Linux you can see present SPD addresses with following commands:

phcoder@sid:~/coreboot/util/autoport$ sudo modprobe i2c-dev
phcoder@sid:~/coreboot/util/autoport$ sudo i2cdetect -l
i2c-0	i2c       	i915 gmbus ssc                  	I2C adapter
i2c-1	i2c       	i915 gmbus vga                  	I2C adapter
i2c-2	i2c       	i915 gmbus panel                	I2C adapter
i2c-3	i2c       	i915 gmbus dpc                  	I2C adapter
i2c-4	i2c       	i915 gmbus dpb                  	I2C adapter
i2c-5	i2c       	i915 gmbus dpd                  	I2C adapter
i2c-6	i2c       	DPDDC-B                         	I2C adapter
i2c-7	i2c       	DPDDC-C                         	I2C adapter
i2c-8	i2c       	DPDDC-D                         	I2C adapter
i2c-9	smbus     	SMBus I801 adapter at 0400      	SMBus adapter
phcoder@sid:~/coreboot/util/autoport$ sudo i2cdetect 9
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-9.
I will probe address range 0x03-0x77.
Continue? [Y/n] y
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          -- -- -- -- -- 08 -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- 24 -- -- -- -- -- -- -- -- -- -- --
30: 30 31 -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: 50 -- -- -- 54 55 56 57 -- -- -- -- 5c 5d 5e 5f
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --

Make sure to replace 9 with whatever bus is marked as smbus. Here in an example you see SPD at address 0x50. Since we've booted with just the module in C0S0, so the first entry in SPD map has to be 0x50. Once you have SPD map your mainboard_get_spd should look something like:

void mainboard_get_spd(spd_raw_data *spd) {
	read_spd (&spd[0], 0x50);
	read_spd (&spd[1], 0x51);
	read_spd (&spd[2], 0x52);
	read_spd (&spd[3], 0x53);
}

You can and should omit lines which correspond to slots not present on your machine.

Note: slot labelling may be missing or unreliable. Use inteltool to see which slot have modules in them.

This way works well if your RAM is socketed. For soldered RAM if you see its SPD, you're in luck and can proceed the same way although you may have to guess some entries due to RAM not being removable.

Most cases of soldered RAM don't have EEPROM chip. In this case you'd have to create fake SPD. Look in inteltool.log. You'll see something like:

/* SPD matching current mode:  */
/* CH0S0  */
00: 92 11 0b 03 04 00 00 09 03 52 01 08 0a 00 80 00
10: 6e 78 6e 32 6e 11 18 81 20 08 3c 3c 00 f0 00 00
20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 65 00
40: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 6d 17
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
/* CH1S0  */
00: 92 11 0b 03 04 00 00 09 03 52 01 08 0a 00 80 00
10: 6e 78 6e 32 6e 11 18 81 20 08 3c 3c 00 f0 00 00
20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 65 00
40: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 6d 17
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00

This is not completely exact represantation of RAM capablities as it lists only the mode currently used and lacks minor info like serial number. Using xxd you can create binary represantation of this SPD:

cat | xxd -r > spd.bin  <<EOF
00: 92 11 0b 03 04 00 00 09 03 52 01 08 0a 00 80 00
10: 6e 78 6e 32 6e 11 18 81 20 08 3c 3c 00 f0 00 00
20: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
30: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 65 00
40: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
50: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
60: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
70: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 6d 17
80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
90: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
a0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
b0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
c0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
d0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
e0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
f0: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
EOF

Then you can place this file into mainboard directory and hook it up into build system by adding following lines to Makefile.inc (creating a new file if needed)

cbfs-files-y += spd.bin
spd.bin-file := spd.bin
spd.bin-type := raw

And then make coreboot actually use this SPD. Following example shows a hybrid situation with one module soldered and another is socketed:

void mainboard_get_spd(spd_raw_data *spd)
{
	void *spd_file;
	size_t spd_file_len = 0;
	/* C0S0 is a soldered RAM with no real SPD. Use stored SPD.  */
	spd_file = cbfs_boot_map_with_leak( "spd.bin", CBFS_TYPE_RAW,
					 &spd_file_len);
	if (spd_file && spd_file_len >= 128)
		memcpy(&spd[0], spd_file, 128);
	/* C0S0 is a DIMM slot.  */
	read_spd(&spd[1], 0x51);
}

If several slots are soldered there are 3 ways of doing things:

  • If all of them are the same use the same file. Don't forget to copy it to all array elements.
  • Use several files (recommended). Name them e.g. spd1, spd2,...
  • Concatenate it into a single file and split into several array elements on runtime. Not recommended

board_info.txt

board_info.txt is a simple text file used to generate wiki page listing supported boards. Some of the info can't be detected automatically.

As this is used only to present information to users then when it matches your board and definitions it is correct.

  • Which package is used for ROM and whether it's socketed, as well as release year. For ROM package refer to http://flashrom.org/Technology and compare it with ROM you found at the beginning of the port. Set ROM package, ROM socketed and other variables mentioned in FIXME.
  • Release year, just go to web and find that information.
  • Category. It's difficult to make difference between desktop and similar categories from inside the computer. Valid categories are:
    • desktop. Desktops and workstations.
    • server. Servers
    • laptop. Laptops.
    • half. Embedded / PC/104 / Half-size boards.
    • mini. Mini-ITX / Micro-ITX / Nano-ITX
    • settop. Set-top-boxes / Thin clients.
    • eval. Devel/Eval Boards
    • sbc. Single-Board computer.
    • emulation. Virtual machines and emulators. May require especial care as they often behave differently from real counterparts.
    • misc. Anything not fitting the categories above. You probably shouldn't use this.

USBDEBUG_HCD_INDEX

Which controller the most easily accessible USB debug port is. On intel 1 is for 00:1d.0 and 2 is 00:1a.0 (yes, it's reversed). See https://www.coreboot.org/EHCI_Debug_Port for more info.

If you're able to use EHCI debug port without setting HCD index manually in config this is correct.

BOARD_ROMSIZE_KB_2048

Default rom size should be detected automatically but sometimes isn't. If yours weren't detected put correct rom size here to serve as sane default when configuring coreboot.

If default ROM size when slecting the board is right one than this value is correct.

DRAM_RESET_GATE_GPIO

When machine is going to S3 in order not to reset the RAM modules, the reset signal must be filtered out from reachin RAM. This is done by powering down a special gate. Most manufacturers put this gate on GPIO 60 but Lenovo is knowon to put it on GPIO 10. If you're able to go to S3 and back than this value is correct.

GNVS

acpi_create_gnvs sets values in GNVS which in turn is used by ACPI for various power-related functions. Normally there is no need to modify it but it makes sense to proofread it.

gfx.ndid and gfx.did

Those describe which video outputs are declared in ACPI tables. Normally there is no need to adjust but if you miss some nonstandard output you can declare it there. Bit 31 is set to indicate presence of the output. Byte 1 is the type and byte 0 is used for disambigution so that ID composed of byte 1 and 0 is unique. Types are

  • 1 = VGA
  • 2 = TV
  • 3 = DVI
  • 4 = LCD

c*_acpower and c*_battery

Which mwait states to match to which ACPI levels. Normally no need to modify unless your device has very special power saving requirements.

install_intel_vga_int15_handler

This is used to configure int15 hook used by vgabios. Parameters 2 and 3 usually shouldn't be modified as vgabios is quite ok to figure panel fit and active output by itself. Last number is the numerical ID of display type. If you don't get any output with vgabios you should try different values for 4th parameter. If it doesn't help try different values for first parameter as well

CMOS options

Due to horrible state of CMOS support in coreboot tree, autoport doesn't support it and this probably won't change until format in the tree improves. If you really care about CMOS options:

  • Create files cmos.layout and cmos.default
  • Enable HAVE_OPTION_TABLE and HAVE_CMOS_DEFAULT in Kconfig

EC (lenovo)

You need to set has_keyboard_backlight (backlit keyboard like X230), has_power_management_beeps (optional beeps when e.g. plugging the cord in) and has_uwb (third MiniPCIe slot) in accordance to functions available on your machine

In rare cases autoport is unable to detect GPE. You can detect it from dmesg or ACPI tables. Look for line in dmesg like

ACPI: EC: GPE = 0x11, I/O: command/status = 0x66, data = 0x62

This means that GPE is 0x11 in ACPI notation. This is the correct value for THINKPAD_EC_GPE. To get the correct value for GPE_EC_SCI you need to substract 0x10, so value for it is 1.

The pin used to wake the machine from EC is guessed. If your machine doesn't wake on lid open and pressing of Fn, change GPE_EC_WAKE.

Keep GPE_EC_WAKE and GPE_EC_SCI in sync with gpi*_routing. gpi*_routing matching GPE_EC_WAKE or GPE_EC_SCI is set to 2 and all others are absent.

If your dock has LPC wires or needs some special treatement you need to fill h8_mainboard_init_dock and add support code to DSDT. See the code for x60, x200 or x201

EC (generic laptop)

Almost any laptop has an embedded controller. In nutshell it's a small low-powered computer specialised on managing power for laptop. Exact functionality differs between macines but of interest to us is mainly:

  • Control of power and rfkill to different component
  • Keyboard (PS/2) interface implementation
  • Battery, AC, LID and thermal information exporting
  • Hotkey support

autoport automatically attempts to restore the dumped config but it may or may not work and may even lead to a hang or powerdown. If your machine stops at Replaying EC dump ... try commenting EC replay out

autoport tries to detect if machine has PS/2 interface and if so calls pc_keyboard_init and exports relevant ACPI objects. If detection fails you may have to add them yourself

ACPI methods _PTS (prepare to sleep) and _WAK (wake) are executed when transitioning to sleep or wake state respectively. You may need to add power-related calls there to either shutdown some components or to add a workaround to stop giving OS thermal info until next refresh.

For exporting the battery/AC/LID/hotkey/thermal info you need to write acpi/ec.asl. For an easy example look into apple/macbook21 or packardbell/ms2290. For information about needed methods consult relevant ACPI specs. Tracing which EC events can be done using dynamic debug

EC GPE needs to be routed to SCI in order for OS in order to receive EC events like "hotkey X pressed" or "AC plugged". autoport attempts to detect GPE but in rare cases may fail. You can detect it from dmesg or ACPI tables. Look for line in dmesg like

ACPI: EC: GPE = 0x11, I/O: command/status = 0x66, data = 0x62

This means that GPE is 0x11 in ACPI notation. This is the correct value for _GPE.

Keep GPE in sync with gpi*_routing. gpi*_routing matching GPE - 0x10 is set to 2 and all others are absent. If EC has separate wake pin then this GPE needs to be routed as well