Re: [coreboot] Why does the Opteron 6328/6220 have two NUMA nodes?
Socket C32 uses a single die per package, G34 is always a dual die MCM, even if it's merely one of the 4 or 8 core SKUs. This is because it needs the dual channel memory controllers from two bulldozer dies in order to support the quad channel memory of G34 platforms.On Feb 2, 2018 8:35 PM, taii...@gmx.com wrote: > > I was under the assumption that they are a single CPU package like their > desktop FX-8100/8300 counterparts, however under linux the 6328 is > reported as two NUMA nodes (CPU 0-3 and 4-7) like the dual CPU 6386 > package is. > > This presents a serious performance issue. > > I am not sure what is reported under the stock BIOS as I am unable to > reflash right now, does anyone know if this is a bug? > > > -- > coreboot mailing list: coreboot@coreboot.org > https://mail.coreboot.org/mailman/listinfo/coreboot -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot
Re: [coreboot] Why does the Opteron 6328/6220 have two NUMA nodes?
On one of my Socket C32 systems with an OEM BIOS an 8 core 4280 CPU uses half as much power and reports only a single NUMA node. -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot
Re: [coreboot] [off topic] Opteron CPU missing chips on the bottom
On 01/31/2018 02:16 PM, Timothy Pearson wrote: -BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Since you've mentioned the remaining lands are also misaligned, I'd throw that CPU away. You currently risk burning out the multiphase regulator and/or attached peripherals (RAM, PCIe), depending on if the lands are now misaligned enough to put voltages where they don't belong. Those lands are very robust; the only way I can think of to chip off one or two would be putting the CPU land-side up and dropping something on it. For more extensive damage, it's possible the CPU was repeatedly overheated at some point in the past as well. The expert has spoken! Thanks for the advice all. I will add a section to the wiki about insisting on a bottom photo for used G34 CPU's. -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot
[coreboot] Why does the Opteron 6328/6220 have two NUMA nodes?
I was under the assumption that they are a single CPU package like their desktop FX-8100/8300 counterparts, however under linux the 6328 is reported as two NUMA nodes (CPU 0-3 and 4-7) like the dual CPU 6386 package is. This presents a serious performance issue. I am not sure what is reported under the stock BIOS as I am unable to reflash right now, does anyone know if this is a bug? -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot
Re: [coreboot] Does the coreboot build system support external patches?
In addition to what Martin suggested they might also consider storing their code in a separate repository and pulling it in as a git submodule into the coreboot/3rdparty/ directory. What's best for them depends on their use case, like if they're mostly adding new files with an API or modifying existing files. On Thu, Feb 1, 2018 at 12:26 PM, Martin Rothwrote: > coreboot uses the site-local directory to allow users to extend the > coreboot build outside of the regular coreboot coreboot files. > Create the coreboot/site-local directory, and add Makefile.inc and Kconfig > files. Both of these files will get picked up by coreboot. > > To run steps at the start of the build, you can create a target that does > whatever is needed, then calls the 'all:' target to do the build. > At the end of the build, you can hook into the 'build_complete::' target > to run whatever finalization steps are needed. > > Martin > > On Thu, Feb 1, 2018 at 10:53 AM, Jay Talbott com> wrote: > >> I got this question from a client yesterday that I didn’t know how to >> answer because I focus primarily on the porting effort (getting coreboot to >> build and boot for their hardware) and let the client figure out how they >> want to manage the resulting code base once the porting effort is complete. >> So I offered to inquire of the coreboot community to see if anybody can >> provide an answer for my client. >> >> >> >> Below is the inquiry that I received: >> >> >> >> I’m looking at how best to structure our coreboot repositories. Does the >> coreboot build system support the concept of external patches? With >> buildroot you can keep your changes is a directory tree separate from the >> FOSS tree. Then during the build step your patches get applied and compiled >> using another separate build tree. >> >> >> >> Thanks, >> >> >> >> - Jay >> >> >> >> Jay Talbott >> Principal Consulting Engineer >> SysPro Consulting, LLC >> 3057 E. Muirfield St. >> Gilbert, AZ 85298 >> (480) 704-8045 >> (480) 445-9895 (FAX) >> jaytalb...@sysproconsulting.com >> >> http://www.sysproconsulting.com >> >> >> >> -- >> coreboot mailing list: coreboot@coreboot.org >> https://mail.coreboot.org/mailman/listinfo/coreboot >> > > > -- > coreboot mailing list: coreboot@coreboot.org > https://mail.coreboot.org/mailman/listinfo/coreboot > -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot
[coreboot] ThinkPad W520 support
I'm happy to announce a working patch for the ThinkPad W520: https://review.coreboot.org/#/c/coreboot/+/23564/ Further details on this mainboard can be found on the wiki: https://www.coreboot.org/Board:lenovo/w520 Kind regards, Nico Rikken -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot
[coreboot] ENE KB3940Q-A1 embedded controller custom firmware
Greetings, Currently working on a port for the hp g7-2247us laptop, which features an ene kb3940q ec, which hopefully should be very similar to the kb3930 ec, which has a datasheet available to the public in a few places. Said similar ec is used in some OLPC devices, as well as some purism devices, and I was hoping someone in the list would have some contacts with those guys so as to be able to use their ec firmware as a bit of a reference design, but the OLPC ec firmware repo has a 'private' submodule which I cannot access and I simply cannot find a repo for the purism ec firmware to reference. Any assistance you could provide on this matter would be greatly appreciated. Marty E. Plummer -- coreboot mailing list: coreboot@coreboot.org https://mail.coreboot.org/mailman/listinfo/coreboot