Tuesday, February 24, 2015

Anniversary .next ^3 and 3 behaviors

Today marks eighteen years in my Intel journey, so I am keeping up the tradition of writing a blog entry on this day, as started with http://vzimmer.blogspot.com/2012/02/anniversary-day.html. The title exponentiates '.next' versus '.next.next.next'-style usage. This makes me eligible for a half-sabbatical, too. Since the last posting I've firmly moved into the new Seattle office. Below is a picture of the facility from across the street. It is located in south Seattle, adjacent to the International District.


I managed to get a cube-with-view, too.


The view also affords me a vantage in case of an invasion https://twitter.com/vincentzimmer/status/565562482383527936



So with a new office and work anniversary afoot, I will continue my blogging tradition. I'll begin with a few thoughts on being an engaged engineer. These sentiments include:

1. Being present

Even before chancing across http://www.forbes.com/sites/ellevate/2015/02/20/why-face-to-face-meetings-are-so-important/ this evening, I have always believed in face-to-face engagements in the workplace. Although I've always worked at a remote site, namely WA-state Intel, I purposely go out of my way to visit the larger sites, such as Intel in Hillsboro, OR, or the headquarters in Santa Clara.

It's imperative to have such meetings early in an engagement so as to build rapport, and then possibly maintain the relationship remotely for subsequent collaboration. Showing up demonstrates that 'you care' and that the other person 'matters.' Nothing can substitute for these gestures.

Also, having badged into five Intel sites across three states during the preceding four-day-week  using plane, train, and automobile, I have recently practiced what I'm preaching on this topic.

2. Saying you don't know

A colleague in Oregon last Thursday suggested I treat this topic, although I've opined on it before. So here goes.

Another behavior I've observed includes reluctance of many technical leader to say "I don't know." Not having the immediate answer is OK. Just commit to following up and supporting the party. Feigning understanding doesn't help your standing with the other party or the overall business goals.

The technology and business landscape changes constantly. Strictly speaking it's impossible 'to know' everything.' But it is possible to keep and open mind, customer-oriented disposition, and desire to 'learn.'  Let an occasional 'I don't know' help fuel the fire of curiosity and pedagogy.

I've touched on the former two items in past blogs, but I have yet to treat the next topic.

3, Ownership

In general, ownership can be powerful in that it highlights the party culpable for furthering a business or technical agenda. Recently, though, I've seen the dark side of ownership. For example, in some discussions people will assert "I own that." This is a variant of the logical fallacy of argument from authority in that the act of ownership is posited in lieu of providing other data, such as results.

Instead of 'ownership', I am more of a fan of 'helping lead,' 'coordinate', 'drive'. What matters is the business result. None of us truly 'own' anything in the company. The upper management, and reductio ad absurdum, the stockholders, 'own' the company. We are entrusted as employees, and especially as technical leaders, to drive a goal.

On the flight Friday morning, a senior manager gave me a nice definition of a leader. Minimally, a leader must 'have followers' and 'make changes.' I'd say that's necessary but not sufficient. Sufficiency may include a qualification of change such that it is 'conducive to the growth of the business,' and followers who 'grow and advance in their careers' as a result of the interaction. Good stuff.

So next time I hear "I own X," I'd rather hear "This is what we've produced with X."

So speaking of X's, let's talk about security and open source.

On the security front, I don't want to say too much other than note that I'll be speaking at CanSecWest (CSW) next month https://cansecwest.com/post/2015-02-01-11:00:00_Speaker_Announcements. The title of my talk, "UEFI, Open Platforms, and the Defender's Dilemma", sums up my thoughts on platform security.

To begin, let's dissect the title. The first portion of the title UEFI provides mechanism, such as many of the technologies found on http://www.tianocore.org and also recent write-up's like http://firmware.intel.com/sites/default/files/resources/A_Tour_Beyond_BIOS_Using_Intel_VT-d_for_DMA_Protection.pdf.

The next portion of the title Open platforms provides a great opportunity to demonstrate a reduction to practice of these technologies, with a goal of demonstrating the design intent of the specifications in bits.

Finally, the defender's dilemma element of the title will review the relative asymmetry of platform defenders to the attackers. The cardinality of the attack presentations at this conference, namely 3 offense UEFI talks, to my 1 defense talk, brings this home. In fact, I represent one of the few defense talks hosted to be found in this conference series.

So let's get to the mapping of our three behaviors to the elements of the title. I will present in person so I get a check-mark on #1 of Being Present without much fan fare. On the topic of saying I don't know, I will not take the posture of being the omniscient guru of firmware. Instead, I want to enjoin the researchers in a dialogue, learning from them as much as I might be teaching. And I may up uttering those 3 humbling words more than I'd like. Finally, I would never claim to "own" platform security. There is a rich community of security professionals with whom I have the honor of working in the UEFI forum, open source community, and as professional colleagues. We all collaborate on providing a solution. I might lead in some venues, such as the UEFI Forum, but I am loathe to ever say I 'own' this topic area.

So that's for CSW next month.

Another application of these 3 behaviors to an X an include open source and open platforms. I've talked in the past about how thing like Intel(R) FSP can support various open source ecosystems and my excitement around Intel(R) Quark-based platforms like Galileo having open sourced their EDK II implementation. I also noted in my Intel Developer Forum 2014 talk that a book was coming. Well, the book Embedded Firmware Solutions: Development Best Practices for the Internet of Things has been published and can acquired via http://www.amazon.com/Embedded-Firmware-Solutions-Development-Practices/dp/1484200713/ or http://www.apress.com/9781484200711.



An interview with the authors was also recently posted, too http://www.se-eng.com/2015/02/authors-of-embedded-firmware-solutions-include-key-industry-architects-of-open-solutions/.

I am a believer in collaboration, too. This book adds Google and Sage to the list of companies with whom I've publicly presented and co-authored material. Examples can be found at https://www.linkedin.com/in/vzimmer and includes AMI, Cisco, Dell, HP, IBM, Insyde, Intel, McAfee, Microsoft, NICTA, Phoenix, SuSE, Toronto.

So how do the three behaviors map to this topic? The easiest to answer is 'being present.' In the earliest days of Intel FSP and this book I had the opportunity to meet my colleagues in Mt. View, San Jose, and San Francisco. In this various venues I helped forge a relationship that evolved remotely, but the pioneering engagement was face-to-face.

And on the topic of "I don't know", I have to cite the spate of libre boot emails I received internally and externally in the wake of announcements like http://www.fsf.org/news/libreboot-x200-laptop-now-fsf-certified-to-respect-your-freedom. I don't know the extent to which the various business partners can open, but I'm always open to learn about the various viewpoints and desires of the ecosystem. But I do believe in more openness and transparency, just as 'open' is in my CSW title and talk, and the book above talks about 'open' solutions.

And finally, I don't 'own' open source firmware and platforms at Intel. I am luckily to have spent the last 15+ years on a team that has been open sourcing reference implementations of industry standards and firmware. We recently posted some elements of http://firmware.intel.com/projects/minnowboard-max to http://www.tianocore.org, such as https://github.com/tianocore/edk2/tree/master/Vlv2DeviceRefCodePkg and https://github.com/tianocore/edk2/tree/master/Vlv2TbltDevicePkg. These packages, and the ability to absorb Intel FSP binaries from http://www.intel.com/fsp, along with the more granular source plus binary builds, shows progress toward more open platforms in this community.

So that's about it for the inaugural blog of 2015. I've marked a work anniversary with a blog three-years-running on this date, I discussed three behaviors, and finally, I mapped those three behaviors to two case studies.

I have been invited to co-present at the Open Compute Summit http://www.opencompute.org/community/events/summit/ocp-us-summit-2015/ next month, too. I should have conjured up a 3rd case study on this topic area of "firmware and Cloud" challenges" to have a 3-3-3 balance in the former paragraph, but work beckons and time to close up this entry.

Have a safe and happy rest of 2015 to all.





Wednesday, December 31, 2014

Regarding a prediction for 2015

Given it is the final day of 2014, I thought that my final blog of 2014 would treat the year.next which commences in a few hours, maybe with another one of my infamous TL:DR-style postings? Specifically, I'd like to build upon missive from Tim Lewis at http://uefi.blogspot.com/2014/10/security-to-eclipse-user-features-as.html wherein he shared his belief that security concerns in the platform firmware will predominate as the top user story of 2015. Given the sampling below from 2014, he may be observing an unenviable trend line.

Regarding that trend, there have been talks at several venues in 2014, such as the Black Hat (BH) talks a few months ago https://www.blackhat.com/docs/us-14/materials/us-14-Kallenberg-Extreme-Privilege-Escalation-On-Windows8-UEFI-Systems.pdf. This talk had an associated CVE http://www.kb.cert.org/vuls/id/552286 and white paper https://www.mitre.org/sites/default/files/publications/14-2221-extreme-escalation.pdf.

In addition, just this week there were a couple of talks featured on UEFI at the Chaos Communications Conference (CCC), including https://www.youtube.com/watch?v=5BrdX7VdOr0 and the associated presentation https://frab.cccv.de/system/attachments/2557/original/AttacksOnUEFI_Slides.pdf
by researchers from Bromium and Mitre, respectively.

The other CCC talk was on EFI and Thunderbolt https://www.youtube.com/watch?v=5BrdX7VdOr0. Glad to hear from the Matthew Garrett question to the presenter near the end of the talk that the option ROM vulnerability could have been mitigated if the Apple platforms implemented UEFI Secure Boot with option ROM verification (i.e., UEFI Driver Loading) enabled.

These security events and broad use of UEFI and edk II-based platforms definitely argue for the need to have additional industry coordination. This is a role that CERT has done admirably, but as Tim noted, they are now joined by the UEFI Security Response Team (USRT) http://www.uefi.org/security. This effort has been followed by a more code base-specific effort, too http://tianocore.sourceforge.net/wiki/Security. I am happy to be involved in both and although my CVSS-Fu isn't too evolved yet, I'm trying my best in writing technical advisories and doing the associated grading.

It is interesting that the references for both BH and CCC  https://www.youtube.com/watch?v=5BrdX7VdOr0 talks above cite the Intel Framework Specifications at www.intel.com and not the superseding UEFI Platform Initialization (PI) specifications http://www.uefi.org/sites/default/files/resources/PI_Spec_1.3.zip. This might contribute to some of the ambiguity seen in the boot script implementations, for example, noted in the CCC talk, since the Intel Framework specifications are frozen after having been contributed to the UEFI Forum. Maybe posting a ZIP of the PI specs and a PDF of the Framework makes the latter more search-engine friendly?

For the CCC talk https://frab.cccv.de/system/attachments/2557/original/AttacksOnUEFI_Slides.pdf, it was good to see that slide 53 mentions the work done with my Intel colleague

      [11] A Tour Beyond BIOS:
      Implementing S3 Resume with EDK2 –JiewenYao and Vincent Zimmer,
      October 2014

Jiewen and I designed features like the Lock Box protocol https://github.com/tianocore/edk2-MdeModulePkg/blob/master/Include/Protocol/LockBox.h to allow for protecting the boot script memory, via SMM or other container, back in 2009. But issues in seeing the protections deployed are compounded by the sheer difficulty in implementing S3 resume in platforms. Specifically, getting the functionality correct and having less-easily testable properties like securing the boot script has proven challenging. As a platform developer the last call you want to hear during product development is from the SQA team, saying "Everything works great, but our S3 resume testing fails once out of 10,000 times."

The above S3 document represents part of an effort to document some of the design intent on edk II source artifacts. We already have architectural intent documents such as http://www.cs.berkeley.edu/~kubitron/courses/cs194-24-S14/hand-outs/SF09_EFIS001_UEFI_PI_TCG_White_Paper.pdf (including the companion talk https://www.youtube.com/watch?v=lZ505uz1TZ4).

Regarding other defenses, in the S3 talk Corey also mentioned the "SMI Transfer Monitor" (STM). Corey's co-presenter Rafal discusses the STM in more detail at http://invisiblethingslab.com/resources/bh09dc/Attacking%20Intel%20TXT%20-%20paper.pdf. This is also mentioned in CCC 2013 talk http://events.ccc.de/congress/2013/Fahrplan/system/attachments/2266/original/Gal_Diskin_-_Virtually_Impossible_-_30C3__release_version_.pdf by former Intel researcher Gal Diskin.on slide 13. Chapter 13 of Grawrock's 2nd edition of Safe Computing http://www.amazon.com/Dynamics-Trusted-Platform-Building-Approach/dp/1934053171 also describes this feature. The gist of the STM is that it represents a usage of the peer-monitor capability in IA32, a hardware feature defined in chapter 34.15 of the Intel Software Development Manual (SDM) https://www-ssl.intel.com/content/www/us/en/processors/architectures-software-developer-manuals.html. There is a nice peer monitor overview at http://virtualizationtechnologyvt.blogspot.com/2011/07/vmx-and-smm-dual-monitor-mode.html. The use of the peer monitor, which is essentially virtualization controls in System Management Mode (SMM), allows for de-privileging today's ring 0 SMM code, such as that built responsive to the infrastructure in volume 4 of the PI 1.3 specification http://www.uefi.org/sites/default/files/resources/PI_Spec_1.3.zip. So when Corey said "where is the STM specification?", he was referring to the firmware convention for loading the STM, interaction with Intel TXT, and how to expose resource lists for which the guest SMM code can access. This class of firmware interaction is the "STM specification" and it is built upon the hardware specifications of the peer monitor virtualization capability treated in the SDM. The SDM is the bible for software-visible parts of IA32, i.e., the 'architecture.' The SDM hasn't changed over the last decade plus, whereas the STM specification has, er, evolved. 

I have spent quite a bit of time on SMM over the last decade plus, from having defined the Framework SMM CIS http://www.intel.com/content/www/us/en/processors/itanium/efi-smm-cis-v09.html infrastructure http://www.google.com/patents/US6848046, from which the present PI 1.3 SMM design has been derived, to collaborating on peer monitor and STM work. I'm a big fan of isolation technologies and trusted computing, but I'll have to pause here for this 2014 entry on this topic. If I were a smarter guy I would follow in the foot-steps of Fermat's Last Theorem and claim to have the final proof of solving the SMM security issues but excuse myself for today by noting that the answer is "too large to fit in the margin" of this blog http://en.wikipedia.org/wiki/Fermat%27s_Last_Theorem.

Just as a document cannot replace well commented codes, having more executable methods for assurance are always preferred. These include efforts such as the CHIPSEC work https://github.com/chipsec/chipsec which I also describe in http://www.uefi.org/sites/default/files/resources/2014_UEFI_Plugfest_04_Intel.pdf and https://www.youtube.com/watch?v=aV1DSF4cwGw. These tests are run after boot and provide an a posteriori view of the platform.

More recent technology includes having security testability in-situ with the platform, versus a CHIPSEC style post-boot test, and reporting the results. This type of test and reporting capability was mentioned in 2013 at http://www.uefi.org/sites/default/files/resources/UEFI_Summerfest_2013_-_Microsoft_Hardware_Security_Test_Interface.pdf and now has a public specification realization at  http://msdn.microsoft.com/en-us/library/windows/hardware/dn879006.aspx (HSTI) This work bridges the semantic gap historically faced by platform builders wherein the various system requirements in a Windows Hardware Certification document, such as http://download.microsoft.com/download/7/0/E/70E74967-B0FE-477A-974F-C1ED16EE31DF/windows8-1-hardware-cert-requirements-system.pdf, can now have a companion reporting channel. Examples include injunctions in System.Fundamentals.Firmware.UEFILegacyFallback, such as

"An OEM may not ship a 64 bit system which defaults to legacy BIOS or loads legacy option ROMs if that system ships with a UEFI-compatible OS. When Secure Boot is Enabled, Compatibility Support Modules (CSM) must NOT be loaded."
These Compatibility Support Module (CSM) behaviors are something that only the platform implementation can deduce and are difficult to detect via a later test, although heuristics like scanning 0xC0000p through 0x100000p in the address space 'could' be done.

The HSTI interface builds upon the Adapter Interface Protocol (AIP) class of structure defined in the UEFI 2.4 specification http://www.uefi.org/sites/default/files/resources/2_4_Errata_A.pdf. Since AIP producers are boot service only, a pre-OS agent needs to query the HSTI interface in order to cull the settings and convey the results into the operating system run time, if needed, in a vendor-specific fashion.

Related work to HSTI that goes beyond mapping to Windows requirements include the ability to report BIOS Integrity Attributes, as described in http://csrc.nist.gov/publications/drafts/800-155/draft-SP800-155_Dec2011.pdf in 3.2.1.1. And in order to surface this information, we defined in http://www.trustedcomputinggroup.org/files/static_page_files/D4268663-1A4B-B294-D05DE87EE90558B6/TCG_EFI_Platform_1_22_Final_-v15.pdf the following:

    typedef struct {
       UINT32 VendorId;
       EFI_GUID ReferenceManifestGuid;
    } TCG_Sp800-155-PlatformId_EventStruct

This structure can be used to reference an off-machine Reference Manifest (RM). The idea is that the GUID in the structure above would be specific to a vendor, model, and firmware instance in a XML-based manifest document on the platform suppliers website. The schemas are defined in http://www.trustedcomputinggroup.org/files/resource_files/73638609-1A4B-B294-D0902CC2A13C8697/Reference_Manifest_Schema_Specification_v2.0.r5.pdf, so now it's a set of work items for the industry to map the RM's to platform attributes, such as mentioned in 155.

Who knows, at some point maybe these reporting and testing efforts will all align?

Well, if you have gotten this far, the 'Didn't Read" portion of "Too Long, Didn't Read" (TL;DR) hasn't been the case for you. I'm not sure whether to reward you for your stoicism or not for wading through this entry.

To summarize, 2014 has shows a spate of security-related activities, both by the attackers (BH, CCC) and the defenders (USRT). The firmware-producing and consuming community is acutely aware of assurance concerns and new testing methodologies have been deployed in '14 and are coming in '15, too, including HSTI. I don't necessarily subscribe to the 'Spy versus Spy' aspect of security, or fall into existential angst that the "Defenders Dilemma" is not solvable. Instead, I see many parties passionate about having robust, usable systems and collaboration on art to achieve those goals. Open source, open documentation, open specifications, open platforms,.... All help in the cause.

And if you're still working these last hours of today, pause and have a coffee.
Also, enjoy a happy and safe new year.

Cheers,

V.Z.兹梅

Saturday, December 20, 2014

"So Long, and Thanks for All the Fish"

Yesterday my last day in DuPont after nearly 18 years, 15 of those in the same cube, here at this WA Intel site. Off to work at the Seattle office in early '15. So long, DuPont, and thanks for the memories. And all of the Fish, although in this case 'Andrew Fish.' Although Fish doesn't work at Intel any longer, the inventor of the "Intel Boot Initiative" (IBI), which became EFI/UEFI, still lives in the South Puget Sound. We broke bread for the final time at our favorite local restaurant.


This is the "R&R" mentioned in the acknowledgements section of the 2nd edition of Beyond BIOS http://www.amazon.com/Beyond-BIOS-Developing-Extensible-Interface/dp/1934053295/, "Also many of this team will recall over ten years of "design discussions" at R&R." I captured Fish on the left and Mark Doran http://newsroom.intel.com/community/intel_newsroom/bios?n=Mark%20S.%20Doran&f=searchAll on the right eating our final Teriyaki bowls. As always, good conversations on firmware and technology. And the picture is also epic in that the duo often introduces itself as "Andrew Fish who invented EFI, and Mark Doran who delivered it to the industry."  Mike Kinney & I held down the other side of the table. I've learned quite a bit working with all three of these colleagues over these many years.

During my final days at Compaq in late 1996 I talked to Intel about joining, and my options were "Join Portland for Xeon, or DuPont for Merced." I was inspired by the thought of joining the 64-bit revolution, so there was no doubt in my mind as to which locale to choose.  Also, having grown up in Texas, the Pacific Northwest was a large unknown, so the distinction between OR and WA didn't even factor into my decision.

And my whole tenure at Intel in DuPont since February 1997 has been boot firmware.  I began with a PC/AT BIOS underneath the Merced System Abstraction Layer (SAL), with the original OS loader interface of SAL_PROC (0x13), as a 64-bit mapping of BIOS 13h.  Same for video mapping of SAL_PROC (0x10), keyboard of SAL_PROC (0x13), etc. The OS community wasn't amused.

As the OS interface evolved to C-based IBI, then EFI, the underlying boot infrastructure changed. I moved from driving some of the SAL-on-BIOS for server (aka "FrankenBIOS"), to the short-lived Workstation Product Division (WPD) where we evolved a C-based boot solution. The latter pioneered art such as the Furball and interposer, which were the predecessor of the Intel Framework Compatibility Support Module (CSM) and CSM16 binary.  We called the former 'furball' because it was something we imagined a native 32-bit (IA32) and 64-bit (Itanium) firmware solution would eventually 'cough up', with a less than pleasant imagery derived from the world of cats.

In 1999 we had evolved the workstation firmware to binary modules called "Plug-In Modules", or PIM's. I recall a fateful meeting with the lead on that effort and my colleague visiting from OR. The conversation went something like 'And we can also put the PIM's into option ROMs, too.' At that point I realized the trajectory was ill-fated since the problem with open platforms and option ROM's isn't factoring code to run in that storage container so much as 'how' to interoperate with the system firmware, namely what API's to expose, where to define them, and how to evolve this art. And to that end, I realized that EFI would win as it was the only man standing to solve this problem.

Specifically, EFI nailed down the interface methodology to the OS loader and option ROM community where the broad interoperability was needed.  Moving the underlying initialization firmware to a new methodology, such as what became Intel Framework and then UEFI PI was valuable, but not the necessary first step.  The first step was to lay the foundation for broad interop in order to have time for the top-level OS and IHV ecosystem to start moving their dependent infrastructure.

At that point in my career when I had that realization, I was lucky enough to have a friend and Intel colleague like Andrew Fish who suggested I have lunch with Mark Doran.  Afterward I on-boarded to the EFI team in 1999 and the rest has been a great 15 run on helping evolve EFI into UEFI, EFI-on-BIOS into Framework and then UEFI PI, and the open source offering from "EFI Sample" to EDK to today's EDK II.  Since then a lot of code and specifications have ensued



with hopefully more to come in 2015+.  Not sure if there is another 15 years into the run.  Although one co-developer of Tiano, the code-name for the underlying Framework (and now PI work) quipped a few years ago "We said Tiano was the '20 year' BIOS replacement solution, so when did you start counting?"

Speaking of 2015+, I had better get back to work if I want to get a head start on the next way of the firmware revolution started in South Puget Sound, WA.

Cheers




Monday, November 17, 2014

Porting to a new architecture

Part of the excitement of working with system firmware includes programming on bare metal. I recall taking an NT Driver class back in 1998 led by some Ex-DEC engineers. The engineer said "programming kernel mode drivers is tricky, but the real magic entails the work of the firmware guys." As an embedded guy working in this area since 1992, all I could say is 'Yes.'

The news about the RISC-V http://riscv.org/ port of coreboot http://review.coreboot.org/#/c/7065/ reminded me of some of the thrills in porting firmware to a new CPU architecture. A possible UEFI port for RISC-V is even mentioned during the technical discussion of this new instruction set architecture https://www.youtube.com/watch?v=vB0FC1DZqUM&feature=youtu.be.

Some of my favorite work includes bring-up for EFI on new CPU's. We started the EFI journey in 1998 working on Itanium and IA32. The Itanium port built upon SAL-A/B as the PI equivalent code, such as memory and CPU initialization http://www.amazon.com/IA-64-Linux-Kernel-Design-Implementation/dp/0130610143.  Simarly, for IA32 we originally booted upon a PC/AT BIOS using techniques like DUET https://github.com/tianocore/edk2/tree/master/DuetPkg.

Over time we moved to native code for the platform initialization as the Framework specifications evolved. These Framework specifications became the UEFI Platform Initialization (PI) specifications, and the code base evolved from EDK http://tianocore.sourceforge.net/wiki/EDK to EDKII http://tianocore.sourceforge.net/wiki/EDK2.

In those early 2000's some fun path-finding was a port to XScale http://en.wikipedia.org/wiki/XScale. I mention that work a bit in the above article and also discuss the port in chapter 7 of  the first edition of Beyond BIOS http://www.amazon.com/Beyond-BIOS-Implementing-Extensible-Interface/dp/0974364908 since Intel was still delivering XScale-based products, although EFI wasn't an official requirement. Instead, I used the XScale port to show the cross-architecture flexibility and scaling of both the specifications and the associated code bases. In the second edition http://www.amazon.com/Beyond-BIOS-Developing-Extensible-Interface/dp/1934053295/, I removed that content and replaced it with Intel IA32 SOC examples.

Ironic to think that I was possibly the first person to do an EFI port to 32-bit ARM, albeit EDK-based and it had a faked-PEI but full DXE.

After that the next excitement occurred with Intel64/EM64T/AMD64/X86_64, or as is now known in the UEFI binding, x64.  See chapter 2 of the UEFI2.4 specification at www.uefi.org. There the challenge was having to enable paging. Itanium also needed paging but we used software TLB handlers and had simple logic to program the TR registers in our SAL, then DXE CPU driver. For x64, we kept PEI to 32-bit since page tables in ROM and / or temporary RAM were tricky and ported DXE, our EFI/UEFI core, to 64-bits.

The crafting of .s/.asm files for a new architecture and getting through the first boot is more exciting than a platform port within a given architecture since there is a sense of pioneering with these foundation components.

Since this pioneering, the ARM community has done the EDKII ports to ARM32 and Aarch64 and they are first class citizens in the standards, including UEFI & PI, and the codebase, such as EDKII.

And although there isn't an official MIPS UEFI binding, there is a port to MIPS 32 at http://sourceforge.net/projects/efi-mips/.

I look forward to seeing more ports to UEFI, including the presently non-UEFI Power8 firmware such as https://github.com/open-power/skiboot.

Like the RISC-V work and ARM ports, I'll have to watch the latter efforts from the sidelines. My day job and extra-credit/after hours budget have enough Intel-architecture based efforts and opportunities to pursue.  Although I have to admit I miss some of the excitement of the port to ISA.Next.


Monday, September 15, 2014

Post-Intel Developer Forum 2014

I would like to thank Tim Lewis for mentioning this blog at http://uefi.blogspot.com/2014/09/new-bios-related-blogs.html. I have dual-posted this item to https://uefidk.com/blog/edkii-fsp-and-other-topics, too, as an experiment. The later community has some interesting code projects and write-up's I reference below.

I spent a few days last week in San Francisco at the Intel Developer Forum (IDF).  My hotel was right off Union Square, thus allowing me to witness of cross-section of San Francisco each morning.






All of the sessions were at the Moscone Convention Center.


My presentation at https://intel.activeevents.com/sf14/connect/sessionDetail.ww?SESSION_ID=1265 now has the material posted https://intel.activeevents.com/sf14/connect/fileDownload/session/08F28DCA2DB2EA88418F93B309D82DE4/SF14_STTS001_102f.pdf.



The audio variant of the material should be posted in a couple of weeks. My talk immediately preceded and also mentioned the session of UBMS https://intel.activeevents.com/sf14/connect/sessionDetail.ww?SESSION_ID=1266 led by Brian Richardson and Mark Doran.

As always, the most exciting part of the event entailed discussion with customers and collaborators both within and outside of Intel. My session was at 9:30am on Thursday and I took a short trip down memory lane, remarking on the two presentations I made in 2003, the first of which entailed how to build out platforms using the Intel (R) Framework specifications http://www.intel.com/content/www/us/en/architecture-and-technology/unified-extensible-firmware-interface/efi-specifications-general-technology.html along with Bob Hart from Insyde and Mike Kinney from Intel, which became the foundation for chapter 7 of the Beyond BIOS books http://www.amazon.com/Beyond-Bios-Implementing-Extensible-Interface/dp/0974364908 and http://www.amazon.com/Beyond-BIOS-Developing-Extensible-Interface/dp/1934053295.

My talk picked up off Brian and Mark's talk from last year https://intel.activeevents.com/sf13/connect/fileDownload/session/DB60155205A8DF5837DA22D0FF90E3A3/SF13_STTS001_100.pdf and built out the progress during the intervening twelve months. I was happy to have Giri Mudusuru and Jiewen Yao from the PC Client Group (PCCG) and my Software and Services Group (SSG) help with questions and answers at the end, too, in addition to serving as co-authors on some of the white papers mentioned later in this blog.

Beyond the FSP-on-EDK2, which is a hybrid of source plus binary, the session also treated on alternate enabling models, including pure open source. Since the session only ran for one hour, with 50 minutes of presentation and 10 minutes of Q&A, we wanted to provide both the attendees and the broader ecosystem an opportunity to 'go deeper' on the subjects treated. As such, beyond the above-listed presentation material, we also note some addition material on slide 27.


The two papers we posted ahead of the talk includes the FSP-on-EDK2 aspect in http://www.uefidk.com/sites/default/files/resources/A_Tour_Beyond_BIOS_Using_the_Intel_Firmware_Support_Package_with_the_EFI_Developer_Kit_II_0.pdf for the initial portion of the third agenda item, and http://www.uefidk.com/sites/default/files/resources/A_Tour_Beyond_BIOS_Implementing_the_TinyQuark_Design.pdf for the second portion of fourth agenda item, respectively. The idea behind the white papers is to 'add words between the Power Point pictures', or elaborate on some of the design intent behind these two engineering efforts.

Another reveal at the talk is the upcoming book from Apress Embedded Firmware Solutions. The alpha posted online http://www.apress.com/9781484200711 already treats the coreboot plus FSP topic, with the final book due early next year. Again, this type of material production aligns with the intent of providing more behind-the-scenes detail and intent of these efforts. Presentations are great to set the tone and direction, and open source repositories must exist in the 'reduction to practice' phase, but having the written word in between the evangelism and the bits helps close the loop for the development community, in my opinion.

Some of the material left of the cutting room floor, in the interest both of time and scope of the presentation, included the recent UEFI payload for coreboot built with EDKII http://www.uefidk.com/develop#accordions_develop-page-9. This was going to follow foil 16 of the presentation were the concept of a 'payload' is introduced, but the inclusion seemed to veer astray of the EDKII/coreboot + FSP tenor of this section.

Here are a couple of the redacted foils that show the flexibility of these open source firmware communities, EDKII, FSP, and coreboot.


and


In the first diagram above you can see that the payload is composed of modules out of EDKII built into a firmware volume. This is the spatial view. We didn't want to mix GPL-based coreboot and EDKII sources, so the only import from the coreboot community was the coreboot memory declarations, or CBMEM, which were also defined in coreboot's libpayload, a BSD-friendly licensed set of sources. The second diagram above shows the flow of the UEFI payload with the coreboot platform 

The ultimate theme of the presentation, though, was to offer customers choice. The imagery used to reinforce this point included the quotation "All Roads Lead to Rome" http://en.wiktionary.org/wiki/all_roads_lead_to_Rome with 'Rome' in this case representing the customer using Intel(R) products and the 'Roads' describing the various enabling options.

Beyond IDF and its technical sessions, though, there was a rare Sergey Brin sighting http://www.theverge.com/2014/9/10/6133589/this-photo-of-sergey-brin-doing-yoga-at-IDF-is-not-strange, although my picture is a bit blurrier

And IDF also featured live music Weds night, namely Weezer http://www.weezer.com/

which was pretty interesting.

While at IDF I also had a chance to work directly with Jiewen, and overseas colleague. We typically interact over email or phone given his home site in the Shanghai, PRC, area and mine in the Seattle, WA, US area. One of the common questions we receive about UEFI Secure Boot UEFI Secure Boot entails Authenticated Variables. This was a topic treated lightly in the implementation document, so Jiewen and I created http://www.uefidk.com/sites/default/files/resources/A_Tour_Beyond_BIOS_Implementing_UEFI_Authenticated_Variables_in_SMM_with_EDKII.pdf to specifically address the SMM-protected variant of this driver posted on http://www.tianocore.og.

On any of these presentations, white papers, and even this blog posting, I look forward to community feedback.

2/15/2015 update - the book cited at the end of the IDF presentation is now for sale http://www.apress.com/9781484200711

Friday, August 15, 2014

ToorCamp redux, Upcoming IDF 2014

To lead off today, I was lucky to have the opportunity to speak at ToorCamp this summer, with my talk titled “Secure boot, network boot, verified boot, oh my” http://toorcamp.toorcon.net/talks/#16 and material posted to https://docs.google.com/file/d/0BxgB4JDywk3MdnRsbnh6NW9rYU0/edit
I especially liked my quotation

A reminder from the KGB school of cipher security: “You never attack the standard, you attack the implementation, including the process.” - Grugq



The locale is pretty remote, namely the western-most portion of the US
My talk was on a Thursday and I was swamped at the office, so I ended up making it a 'day trip' from Tacoma to Neah Bay.

Like 2012, the talks were all hosted in the dome

including the couch for chatting with the host

I have to admit that the speaker on bio-hacking and transhumanism didn't convince me to get an RFID injection into my hand this trip.

I gave my talk and followed up with some face to face discussions for a couple hours afterward. Invariably the question of key revocation came up as a question in response to the Secure Boot discussion. Then I scurried to the nearby beach, snapped a photo of the sunset, and then headed back home. Luckily the local tribesman selling smoked salmon hadn't closed up shop, so I picked up a few packages of the same and beat a hasty retreat.

Saying goodbye to the Makah reservation


and the hackers at Hobuck Beach



The next journey is the Intel Developer Forum in San Francisco.  My upcoming talk is “Firmware Flexibility  using Intel Firmware Support Package,” Talk STTS001, Intel Developer Forum, San Francisco, September 11, 2014 https://intel.activeevents.com/sf14/connect/sessionDetail.ww?SESSION_ID=1265 This should provide a deep dive responsive to the https://intel.activeevents.com/sf13/connect/fileDownload/session/DB60155205A8DF5837DA22D0FF90E3A3/SF13_STTS001_100.pdf presentation last year, along with a few other updates on ecosystems and open source. Drop me a line if you're in SF at this time.

On other progress, my issued US Patents continue to climb, albeit slowly. Now that I've crossed the 300 mark (303 this week for US Patent Families and 823 for INPADOC) maybe I'll get to join the list of 'Prolific Inventors' at http://en.wikipedia.org/wiki/List_of_prolific_inventors. I am curious how to site confirms the assertion: "However, this table currently has an arbitrary cut-off limit for inclusion of 300 patent families. This is purely for practical reasons – there are 81 inventors throughout history with more than 300 utility patent families, but tens of thousands of inventors with more than 15 patents."  Hmmm.

This summer has also witnessed a flurry of presentations on attacks against UEFI implementations, including http://www.mitre.org/sites/default/files/publications/14-2221-extreme-escalation-presentation.pdf. This reminds me of the importance of the Grugq quote above on 'implementation' and underscores the value of work like Chipsec https://github.com/chipsec/chipsec I mentioned in http://www.uefi.org/sites/default/files/resources/2014_UEFI_Plugfest_04_Intel.pdf and other developer guidance, such as 'best practices' in pages 34-35 of http://www-inst.eecs.berkeley.edu/~cs194-24/sp13/hand-outs/SF09_EFIS001_UEFI_PI_TCG_White_Paper.pdf, But there are many additional things we can do with respect to testing, guidance, and instances of best-practices on http://tianocore.sourceforge.net/wiki/EDK2. Speaking of edk2 and security practices, I'm happy to see a reference implementation of a signed capsule update implementation, including the https://svn.code.sf.net/p/edk2/code/trunk/edk2/SecurityPkg/Library/DxeRsa2048Sha256GuidedSectionExtractLib/ support code.

Another exciting open source action is the release of seL4 kernel http://sel4.systems/ and the Isabelle proofs. I mentioned this effort in http://vzimmer.blogspot.com/2013/12/better-living-through-tools.html and since that publication https://github.com/seL4/seL4 has gone live. Gernot and the NICTA guys are impressive. I was happy to see my ex-NICTA collaborator Leonid posted our driver synthesis paper http://www.nicta.com.au/pub?doc=7690 to NICTA's website http://ssrg.nicta.com.au/projects/TS/drivers/synthesis/.  

Regarding the latter paper, this is as close as I can get to a refereed conference, it would seem, as Mike Rothman and I were bounced from LISA '14. We posted the rejected manuscript at https://uefidk.com/sites/default/files/resources/uefi-manageability-security-white-paper.pdf.  Since I'm not an academic but an ordinary Joe who has been slogging away in industry for the last 20+ years, I cannot understand the publish-or-perish or other metrics around papers http://blogs.lse.ac.uk/impactofsocialsciences/2014/04/23/academic-papers-citation-rates-remler/ in academia. I see the value of peer review and appreciate the written-word, along with open source, to scale pedagogy and advocacy.  

Social media never ceases to fascinate me.

Pageviews by Countries

Graph of most popular countries among blog viewers
EntryPageviews
Malaysia
2

Pageviews by Browsers

EntryPageviews
Chrome
2 (100%)
Image displaying most popular browsers

Pageviews by Operating Systems

EntryPageviews
Macintosh
2 (100%)
Image displaying most popular platforms

represents the latest access to this blog.  Someone reading this blog while running Google Chrome on a Macintosh in Malaysia.  Fascinating.

Or on Twitter https://twitter.com/vincentzimmer when I get a re-tweet or message from famous mathematicians
 retweeted your Retweet
Jun 28: 
This is a geometry joke.

Great stuff.

1/7/2015 update -
A friend of mine from Houston just pointed out an update to http://en.wikipedia.org/wiki/List_of_prolific_inventors.
I'm now on the list, and not even the 'bottom-most' entry (and among the youngest who owns up to his/her age).
Vincent Zimmer312 USA8481970-Computer software and firmware[181][182]