diff options
| author | Francis Rowe <info@gluglug.org.uk> | 2015-07-18 19:31:16 (EDT) |
|---|---|---|
| committer | Francis Rowe <info@gluglug.org.uk> | 2015-09-20 04:06:27 (EDT) |
| commit | d1bb5d6fcc75f07ca6a58b5536906980aff9d253 (patch) | |
| tree | 7821900fe16ece180a021957371a7c0b798922dd /docs/hcl/r500.html | |
| parent | 919af0e6b7cbff2171d370a0b830f2b9d740410b (diff) | |
| download | libreboot-d1bb5d6fcc75f07ca6a58b5536906980aff9d253.zip libreboot-d1bb5d6fcc75f07ca6a58b5536906980aff9d253.tar.gz libreboot-d1bb5d6fcc75f07ca6a58b5536906980aff9d253.tar.bz2 | |
New board: ThinkPad R500 (experimental)
The ich9deblob and ich9gen utilities were modified, so that they
support reading and/or writing descriptor images where the GbE
region is not defined. These utilities were also re-factored
and tidied up a bit.
A quick was noticed during the course of this work, in that
Compenent 1 Density was being set to 8MiB constantly, even
on systems with 4MiB flash chips. Component 2 Density was
set statically to 2MiB. ich9gen now sets both to 4MiB or 8MiB,
depending on whether building the descriptor for a 4MiB or
8MiB ROM image.
There are still some ACPI bugs (see docs/hcl/r500.html), which
will have to be fixed upstream. TODO: get hw reg dumps from
a factory R500, and compare with the X200 or T400 dumps.
Diffstat (limited to 'docs/hcl/r500.html')
| -rw-r--r-- | docs/hcl/r500.html | 299 |
1 files changed, 299 insertions, 0 deletions
diff --git a/docs/hcl/r500.html b/docs/hcl/r500.html new file mode 100644 index 0000000..434aac7 --- /dev/null +++ b/docs/hcl/r500.html @@ -0,0 +1,299 @@ +<!DOCTYPE html> +<html> +<head> + <meta charset="utf-8"> + <meta name="viewport" content="width=device-width, initial-scale=1"> + + <style type="text/css"> + @import url('../css/main.css'); + </style> + + <title>ThinkPad R500</title> +</head> + +<body> + + <div class="section"> + <h1 id="pagetop">ThinkPad R500</h1> + + <p> + It is believed that all or most R500 laptops are compatible. + See notes about <a href="../install/r500_external.html#cpu_compatibility">CPU compatibility</a> for + potential incompatibilities. + </p> + + <p> + There are two possible flash chip sizes for the R500: 4MiB (32Mbit) or 8MiB (64Mbit). + This can be identified by the type of flash chip below the palmrest: 4MiB is SOIC-8, 8MiB + is SOIC-16. + <b> + NOTE: this paragraph is being treated with contempt. When an R500 was disassembled, it didn't look + like there was an extra place for SOIC-16. It's highly likely that these laptops only have SOIC-8 (4MiB) + flash chips. For now, libreboot will distribute 8MiB images just in case. If it is found later on + that no 8MiB (SOIC-16) chips exist on the R500, then libreboot will cease to distribute 8MiB ROM images + for this laptop. It is only said that the R500 has 4MiB or 8MiB, for now, since this is the case + on other GM45 thinkpads that are supported in libreboot. + </b> + </p> + + <p> + <b>The R400 laptops come with the ME (and sometimes AMT in addition) before flashing libreboot. Libreboot disables and removes it + by using a modified descriptor: see <a href="gm45_remove_me.html">gm45_remove_me.html</a></b> (contains notes, plus + instructions) + </p> + + <p> + Flashing instructions can be found at <a href="../install/index.html#flashrom">../install/index.html#flashrom</a> + </p> + + <p> + <b> + NOTE: This board is unsupported in libreboot 20150518. To use it in libreboot, for now, + you must build for it from source using the libreboot git repository. + </b> + </p> + + <p> + <a href="index.html">Back to previous index</a>. + </p> + </div> + + <div class="section"> + + <h2 id="compatibility_noblobs">Compatibility (without blobs)</h2> + + <div class="subsection"> + <h3 id="hwvirt">Hardware virtualization (vt-x)</h3> + <p> + The R400, when run without CPU microcode updates in coreboot, currently kernel panics + if running QEMU with vt-x enabled on 2 cores for the guest. With a single core enabled + for the guest, the guest panics (but the host is fine). Working around this in QEMU + might be possible; if not, software virtualization should work fine (it's just slower). + </p> + <p> + On GM45 hardware (with libreboot), make sure that the <i>kvm</i> and <i>kvm_intel</i> kernel modules + are not loaded, when using QEMU. + </p> + <p> + The following errata datasheet from Intel might help with investigation: + <a href="http://download.intel.com/design/mobile/specupdt/320121.pdf">http://download.intel.com/design/mobile/specupdt/320121.pdf</a> + </p> + </div> + + </div> + + <div class="section"> + + <h1 id="lcd_compatibility">LCD compatibly</h1> + <p> + Not all LCD panels are known to be compatible yet. See <a href="gm45_lcd.html">gm45_lcd.html</a>. + </p> + + </div> + + <div class="section"> + + <p> + The R500 is almost identical to the X200, code-wise, but there are some hardware differences. See <a href="x200.html">x200.html</a>. + </p> + + </div> + + <div class="section"> + + <h2 id="regdumps">Hardware register dumps</h2> + + <p> + The coreboot wiki <a href="http://www.coreboot.org/Motherboard_Porting_Guide">shows</a> + how to collect various logs useful in porting to new + boards. Following are outputs from the R500: + </p> + + <ul> + <li> + Lenovo BIOS 3.13 (EC firmware 1.06): + <ul> + <li><a href="hwdumps/r500/">hwdumps/r500/</a></li> + </ul> + </ul> + + </div> + + <div class="section"> + + <h1 id="issues">Issues</h1> + <h2> + False report of overheating, automatic shut down + </h2> + <p> + When attempting to boot Trisquel 7 live USB (GNOME), the following error appears and then + the system abruptly shuts down: + <b>thermal thermal_zone1: critical temperature reached(120 C),shutting down</b>. + </p> + <p> + This is false. When booting with <b>acpi=off</b>, xsensors shows no overheating during a stress test. + The system does not feel hot, nor does anything smell like it's burning. + </p> + <p> + This is most likely caused by an ACPI bug in coreboot, which will have to be investigated. Grep for those things, + comparing factory/libreboot (iasl -d or acpidump): + </p> +<pre> + Return (C2K(\_SB.PCI0.LPCB.EC.TMP0)) +TMP0, 8, /* Thermal Zone 0 temperature */ +</pre> + <h2>Cardbus slot didn't work when tested</h2> + <p> + Investigate. + </p> + + </div> + + <div class="section"> + + <h1>Descriptor differences</h1> + + <p> + The <i>ich9gen</i> and <i>ich9deblob</i> utilities were modified, + to reflect these differences. + </p> + + <h2>Component 1 Density</h2> + +<pre> +- descriptorStruct.componentSection.flcomp.component1Density = 0x4; ++ descriptorStruct.componentSection.flcomp.component1Density = 0x3; +</pre> + + <p> + Read page 848 in the ICH9 datasheet, linked to from <a href="gm45_remove_me.html#flash_descriptor_region">gm45_remove_me.html#flash_descriptor_region</a>. + This doesn't break anything, but in the process of debugging descriptor differences on the R500, it was found that this + config option isn't being modified in libreboot, for different size ROM images. 4MiB ROM images still contain 0x4 for component1Density. + Per datasheets, 0x4 (100) is 8MiB, and 0x3 (011) is 4MiB. <b>This should be fixed!</b> + </p> + <p> + It was 0x3 for this test, because the R500 that was used to create this report + had a 4MiB SOIC-8 flash chip. + </p> + + <h2>flReg1.LIMIT</h2> + +<pre> +- /* descriptorStruct.regionSection.flReg1.LIMIT = 0x07ff; */ ++ /* descriptorStruct.regionSection.flReg1.LIMIT = 0x03ff; */ +</pre> + + <p> + Ignore this. This is not used at all, and is instead automatically set, depending on + the targetted ROM image size, both in ich9gen and ich9deblob. 0x7ff means 8MiB, and 0x3ff means + 4MiB. flReg1 is for the BIOS region. Simply speaking, this is defining the final 4KiB section + of the ROM image, where the BIOS region ends. + </p> + <p> + It was 0x3ff for this test, because the R500 that was used to create this report + had a 4MiB SOIC-8 flash chip. + </p> + + <h2>Onboard gigabit ethernet NIC is disabled</h2> + +<pre> +- descriptorStruct.ichStraps.ichStrap0.integratedGbe = 0x1; +- descriptorStruct.ichStraps.ichStrap0.lanPhy = 0x1; ++ descriptorStruct.ichStraps.ichStrap0.integratedGbe = 0x0; ++ descriptorStruct.ichStraps.ichStrap0.lanPhy = 0x0; +</pre> + + <p> + Most GM45 laptops (e.g. X200, T400, T500, R400) have the <b>Intel 82567LM</b> + integrated gigabit NIC. + </p> + <p> + On the R500, a <b>Broadcom BCM5787M</b> NIC is present. To make this work, + the change above must be made for the R500 descriptor. + </p> + + <h1 id="nogbe">No Gbe region!</h1> + + <p> + Not shown in the diffs above: + </p> + +<pre> +Original: Descriptor start block: 00000000 ; Descriptor end block: 00000000 +Original: BIOS start block: 00200000 ; BIOS end block: 003ff000 +Original: ME start block: 00001000 ; ME end block: 001f7000 +Original: GBe start block: 00fff000 ; GBe end block: 00000000 +Original: Platform start block: 001f8000 ; Platform end block: 001ff000 +</pre> + + <p> + As explained above, this laptop uses a Broadcom NIC, which means that + the Gbe region does not and <i>should not</i> exist, since this is for the + Intel NIC only. + </p> + <p> + In the output above, Gbe starts at fff and ends at 000. Base 1FFF or FFF, and limit 0, + means that the region is disabled. + </p> + <p> + In the output above, the ME region is 4KiB larger than on other GM45 systems that have + a Gbe region. This accounts for the lack of a Gbe region. + </p> + + <p> + As part of this effort, ich9gen/ich9deblob/demefactory will all be + modified to account for the differences above. + </p> + + </div> + + <div class="section"> + + <p> + Copyright © 2015 Francis Rowe <info@gluglug.org.uk><br/> + Permission is granted to copy, distribute and/or modify this document + under the terms of the GNU Free Documentation License, Version 1.3 + or any later version published by the Free Software Foundation; + with no Invariant Sections, no Front-Cover Texts, and no Back-Cover Texts. + A copy of the license can be found at <a href="../gfdl-1.3.txt">../gfdl-1.3.txt</a> + </p> + + <p> + Updated versions of the license (when available) can be found at + <a href="https://www.gnu.org/licenses/licenses.html">https://www.gnu.org/licenses/licenses.html</a> + </p> + + <p> + UNLESS OTHERWISE SEPARATELY UNDERTAKEN BY THE LICENSOR, TO THE + EXTENT POSSIBLE, THE LICENSOR OFFERS THE LICENSED MATERIAL AS-IS + AND AS-AVAILABLE, AND MAKES NO REPRESENTATIONS OR WARRANTIES OF + ANY KIND CONCERNING THE LICENSED MATERIAL, WHETHER EXPRESS, + IMPLIED, STATUTORY, OR OTHER. THIS INCLUDES, WITHOUT LIMITATION, + WARRANTIES OF TITLE, MERCHANTABILITY, FITNESS FOR A PARTICULAR + PURPOSE, NON-INFRINGEMENT, ABSENCE OF LATENT OR OTHER DEFECTS, + ACCURACY, OR THE PRESENCE OR ABSENCE OF ERRORS, WHETHER OR NOT + KNOWN OR DISCOVERABLE. WHERE DISCLAIMERS OF WARRANTIES ARE NOT + ALLOWED IN FULL OR IN PART, THIS DISCLAIMER MAY NOT APPLY TO YOU. + </p> + <p> + TO THE EXTENT POSSIBLE, IN NO EVENT WILL THE LICENSOR BE LIABLE + TO YOU ON ANY LEGAL THEORY (INCLUDING, WITHOUT LIMITATION, + NEGLIGENCE) OR OTHERWISE FOR ANY DIRECT, SPECIAL, INDIRECT, + INCIDENTAL, CONSEQUENTIAL, PUNITIVE, EXEMPLARY, OR OTHER LOSSES, + COSTS, EXPENSES, OR DAMAGES ARISING OUT OF THIS PUBLIC LICENSE OR + USE OF THE LICENSED MATERIAL, EVEN IF THE LICENSOR HAS BEEN + ADVISED OF THE POSSIBILITY OF SUCH LOSSES, COSTS, EXPENSES, OR + DAMAGES. WHERE A LIMITATION OF LIABILITY IS NOT ALLOWED IN FULL OR + IN PART, THIS LIMITATION MAY NOT APPLY TO YOU. + </p> + <p> + The disclaimer of warranties and limitation of liability provided + above shall be interpreted in a manner that, to the extent + possible, most closely approximates an absolute disclaimer and + waiver of all liability. + </p> + + </div> + +</body> +</html> |