Re: [TF-A] Proposal: TF-A to adopt hand-off blocks (HOBs) for information passing between boot stages
Le mer. 19 mai 2021 à 23:51, Jeremy Linton via TF-A < tf-a@lists.trustedfirmware.org> a écrit :
On 5/18/21 8:59 PM, Madhukar Pappireddy via TF-A wrote:
Hi,
I tried to summarize the discussions in the previous TF-A tech forum
regarding the proposal to adopt Hand-off Blocks (HOBs) for passing information along the boot chain. I am certain I could not capture all suggestions/concerns brought up during the call. I apologize if I missed and/or misinterpreted any comments and would appreciate it if everyone could share their thoughts in response to this email thread.
The idea is to share information to other boot phases:
Dynamic information: Created during runtime. Shared in the form of a
chain of blobs(built as a linked list of C structure objects i.e., HOB list).
Static information: Known at compile time. Historically, shared through
the use of Device Tree/ACPI tables
Both the above requirements are common in many ecosystems and need to
co-exist.
There are broadly 3 problems to solve:
- Format of HOB structures: It looks like the consensus is that we
could use existing mechanisms for this (BL_AUX_PARAM in TF-A or bloblist in u-boot).
- Identification of HOB list entries: There is a debate about whether
tags would suffice or if the HOB list producer and consumer would depend on UUID/GUIDs for identifying a specific HOB structure. Another suggestion was to use a hybrid approach. Reserve a single tag ID for identifying/constructing a HOB structure that further leverages UUID based identifier. This way, the generic HOB list doesn't need to support UUIDs and can work with tags.
- The design contract for the static interface between two boot phases:
The problem at hand is whether to pass a pointer to a HOB list or a device tree blob through the general-purpose registers for configuration hand-off between two boot phases. Some proposals that came up:
> Proposal 1: Always pass a pointer to the device tree blobthrough the GP register and capture the pointer to the HOB list as a property of a node that is uniquely identifiable by the downstream boot phase. This needs to define a device tree binding such that producer and consumer agree on the information passed.
Using DT to pass platform info at this level is sort of crazy on an ACPI machine which won't have native DTs. Meaning there is an additional level of unnecessary indirection that needs to be converted back into a format which can be utilized by AML and other parts of the ACPI stack.
I would love the BL33 becomes a product maker (using a board vendor product) decision: using EDK2, U-Boot, LinuxBoot or even Xen directly as BL33. If this is a goal , then there will be a need for a translation layer between the firmware framework (tfa, core boot, uboot SPL) formats and the non secure firmware format. If this is not a goal then I agree that passing a DT in an ACPI system is suboptimal. That said there is a UEFI PI HOB that is a DT container and used in existing systems. And conversely, in systems with complex MDIO or other hardware without ACPI representation (serdes), it does not make sense to bring ACPI stuff. The easiest path is to not give a choice of BL33 and have a format for each “realm”: DT or ACPI.
Its also helpful to look at what has become of the rpi4 uefi port, where the DT is actually dynamic (or provided by the user, complete with HAT overlays) and fed into the lower level firmware and propagated up through the system. The result has frequently been subtle bugs or boot failures because the DT provided by the rpi foundation as part of their low level firwmare+kernel stack is modified by their low level firwmare, and it differs from the DT in mainline linux. So more than once we have discovered that there isn't a single DT that can be boot both the current firmware and a current mainline linux. In the past there was even an option to use one DT for the firmware and an entirely seperate one for linux, but that was removed when it bacame apparent you couldn't have the lower level firmwaer say modifying MMIO windows for the PCIe subsystem and not propogate that into the other user supplied DT.
ACPI shows that firmware provided hardware description makes life easier. That’s why Arm SystemReady-IR states that it should be the case for DT too. RPI4 and beaglebone have had different strategies and each with good and bad things. The Linaro Device Tree Evolution open project is tackling diverse aspects of DT including proper handling hats, and other runtime adaptations.
At this point the uefi firmware on the rpi will provide a DT if asked, but it doesn't parse it nor does anything in the AML. Instead everything is done directly against the hardware, even when the HW registers aren't well documented.That is in large parts because the AML requirements are a lot different than what is provided in the DT.
> Proposal 2: Pass a pointer to a generic container through the GPregister that can be interpreted appropriately by both boot loaders(i.e., producer and consumer of the boot info). This container can either be a dtb or a HOB list which can be simply inferred by checking for a magic header that indicates if the buffer appears to be a flattened device tree.
> One another concern that was brought up offline is to make surewe don't break current design contracts between various boot loader phases in TF-A. Many of the general-purpose registers have a designated purpose such as to share configurations between BL images( such as firmware config dtb, SoC config dtb, Non trusted firmware config dtb, memory layout, entry point info, etc.).
If I am not mistaken, a single design may not fit the needs of every
segment(client, Infra, embedded) and the forum is open to solutions tailored for individual segments. Joanna will be sending a follow up email with more information about future TF-A tech forums that serves as a platform for further discussions.
Thanks, Madhukar
-----Original Message----- From: TF-A tf-a-bounces@lists.trustedfirmware.org On Behalf Of Joanna
Farley via TF-A
Sent: Sunday, May 16, 2021 5:19 AM To: Okash Khawaja okash.khawaja@gmail.com; Simon Glass <
sjg@chromium.org>
Cc: Harb Abdulhamid OS abdulhamid@os.amperecomputing.com; Boot
Architecture Mailman List boot-architecture@lists.linaro.org; tf-a@lists.trustedfirmware.org; Ed Stuber edstuber@amperecomputing.com; Arjun Khare akhare@amperecomputing.com; U-Boot Mailing List < u-boot@lists.denx.de>; Paul Isaac's paul.isaacs@linaro.org; Ron Minnich rminnich@google.com; Moe Ammar moe@amperecomputing.com
Subject: Re: [TF-A] Proposal: TF-A to adopt hand-off blocks (HOBs) for
information passing between boot stages
Apologies I failed with the recording. Manish/Madhu will reply early
next week with the slides and some notes to help with a follow up session which we hope to hold this Thursday. Invite and agenda will also be sent out early next week.
Thanks
Joanna
On 14/05/2021, 13:30, "TF-A on behalf of Okash Khawaja via TF-A" <
tf-a-bounces@lists.trustedfirmware.org on behalf of tf-a@lists.trustedfirmware.org> wrote:
Hi, Do we have slides and video from last week's discussion? Thanks, Okash On Wed, May 5, 2021 at 11:52 PM Simon Glass via TF-A <tf-a@lists.trustedfirmware.org> wrote: > > Hi Harb, > > Thanks for the idea. I am still not completely sure what benefitUUID provides to an open project. I'd like to propose something different, more in the spirit of open collaboration. I also worry that the word 'standard' seems to be a synonym for UUIDs, UEFI, etc., i.e. enabling/preferring closed-source firmware and the continued decline of open-source projects. It really should not be.
> > So I suggest: Use simple integer IDs and reserve some area for'private' use. If you want to collaborate across projects outside your company, you either need to allocate a 'public' ID or agree privately between the parties which private ID to use.
> > This means that the default and easiest option is forcollaboration and a public ID, with private ones (whose purpose may be secret) reserved just for private use.
> > Regards, > Simon > > On Wed, 5 May 2021 at 11:42, Harb Abdulhamid OS <abdulhamid@os.amperecomputing.com> wrote:
>> >> Hey Folks, >> >> We wanted to put out a middle-ground proposal to help guide thediscussion on the call tomorrow.
>> >> >> >> A proposal that we have been discussing offline involvesreserving a single tag ID for the purpose of construction UEFI PI HOB List structure, and that tag would be used to identify a HOB-specific structure that does leverage UUID based identifier. This will eliminate the burden of having to support UUID as the tag, and this enables projects that require UUID based identifiers for the broad range of HOB structures that need to be produced during the booting of the platform. Once we have a tag for a HOB list, this will enable various HOB producers that can add/extend the HOB list in TF-A code (or even pre-TF-A code), with a HOB consumer for that UUID/GUID on the other side (i.e. whatever the BL33 image is booting on that platform).
>> >> >> >> Essentially, the idea is if someone would like to support HOBstructures in a standard way using TF-A, they would wrap it up in a BL_AUX_PARAM/BLOB structure (whatever the group decides) and the way we identify the structure as a HOB list is with this new reserved tag.
>> >> >> >> Hopefully that makes sense and less contentious. Look forwardto discuss this further on the call.
>> >> >> >> Thanks, >> >> --Harb >> >> >> >> From: Manish Pandey2 <Manish.Pandey2@arm.com> >> Sent: Friday, April 30, 2021 8:14 AM >> To: François Ozog <francois.ozog@linaro.org> >> Cc: Simon Glass <sjg@chromium.org>; Julius Werner <jwerner@chromium.org>; Harb Abdulhamid OS < abdulhamid@os.amperecomputing.com>; Boot Architecture Mailman List < boot-architecture@lists.linaro.org>; tf-a@lists.trustedfirmware.org; U-Boot Mailing List u-boot@lists.denx.de; Paul Isaac's < paul.isaacs@linaro.org>; Ron Minnich rminnich@google.com
>> Subject: Re: [TF-A] Proposal: TF-A to adopt hand-off blocks(HOBs) for information passing between boot stages
>> >> >> >> Hi All, >> >> >> >> Please find invite for next TF-A Tech Forum session to continueour discussions on HOB implementation, feel free to forward it to others.
>> >> >> >> The next TF-A Tech Forum is scheduled for Thu 6th May 2021 16:00– 17:00 (BST).
>> >> >> >> Agenda: >> >> Discussion Session: Static and Dynamic Information Handling inTF-A
>> >> Lead by Manish Pandey and Madhukar Pappireddy >> >> · There is ongoing mailing lists discussion[1] relatedwith adopting a mechanism to pass information through boot stages.
>> >> The requirement is two-fold: >> >> 1. Passing static information(config files) >> >> 2. Passing dynamic information (Hob list) >> >> In the upcoming TF-A tech forum, we can start with a discussionon dynamic information passing and if time permits, we can cover static information passing. The purpose of the call is to have an open discussion and continue the discussion from the trusted-substrate call[2] done earlier. We would like to understand the various requirements and possible ways to implement it in TF-A in a generalized way so that it can work with other Firmware projects.
>> >> >> >> The two specific item which we would like to discuss are: >> >> 1. HOB format: TF-A/u-boot both has an existing bloblistimplementation, which uses tag values. Question, can this be enhanced to use hybrid values(Tag and UUID) both?
>> >> 2. Standardization on Physical register use to pass base ofHoB data structure.
>> >> References: >> >> [1]https://lists.trustedfirmware.org/pipermail/tf-a/2021-April/001069.html
>> >> [2]https://linaro-org.zoom.us/rec/share/zjfHeMIumkJhirLCVQYTHR6ftaqyWvF_0klgQnH... Passcode: IPn+5q%
>> >> >> >> Thanks >> >> >> >> Joanna >> >> >> >> You have been invited to the following event. >> >> TF-A Tech Forum >> >> When >> >> Every 2 weeks from 16:00 to 17:00 on Thursday United Kingdom Time >> >> Calendar >> >> tf-a@lists.trustedfirmware.org >> >> Who >> >> • >> >> Bill Fletcher- creator >> >> • >> >> tf-a@lists.trustedfirmware.org >> >> more details » >> >> >> >> We run an open technical forum call for anyone to participateand it is not restricted to Trusted Firmware project members. It will operate under the guidance of the TF TSC.
>> >> >> >> Feel free to forward this invite to colleagues. Invites are viathe TF-A mailing list and also published on the Trusted Firmware website. Details are here: https://www.trustedfirmware.org/meetings/tf-a-technical-forum/
>> >> >> >> Trusted Firmware is inviting you to a scheduled Zoom meeting. >> >> >> >> Join Zoom Meeting >> >> https://zoom.us/j/9159704974 >> >> >> >> Meeting ID: 915 970 4974 >> >> >> >> One tap mobile >> >> +16465588656,,9159704974# US (New York) >> >> +16699009128,,9159704974# US (San Jose) >> >> >> >> Dial by your location >> >> +1 646 558 8656 US (New York) >> >> +1 669 900 9128 US (San Jose) >> >> 877 853 5247 US Toll-free >> >> 888 788 0099 US Toll-free >> >> Meeting ID: 915 970 4974 >> >> Find your local number: https://zoom.us/u/ad27hc6t7h >> >> >> >> ________________________________ >> >> From: François Ozog <francois.ozog@linaro.org> >> Sent: 08 April 2021 16:50 >> To: Manish Pandey2 <Manish.Pandey2@arm.com> >> Cc: Simon Glass <sjg@chromium.org>; Julius Werner <jwerner@chromium.org>; Harb Abdulhamid OS < abdulhamid@os.amperecomputing.com>; Boot Architecture Mailman List < boot-architecture@lists.linaro.org>; tf-a@lists.trustedfirmware.org < tf-a@lists.trustedfirmware.org>; U-Boot Mailing List u-boot@lists.denx.de; Paul Isaac's paul.isaacs@linaro.org; Ron Minnich rminnich@google.com
>> Subject: Re: [TF-A] Proposal: TF-A to adopt hand-off blocks(HOBs) for information passing between boot stages
>> >> >> >> Hi >> >> >> >> here is the meeting recording: >> >>https://linaro-org.zoom.us/rec/share/zjfHeMIumkJhirLCVQYTHR6ftaqyWvF_0klgQnH... Passcode: IPn+5q%z
>> >> >> >> I am really sorry about the confusion related to the meetingtime. I have now understood: the Collaborate portal uses a specific calendar which is tied to US/Chicago timezone while the actual Google Calendar is tied to Central Europe timezone. I am going to drop the Collaborate portal and use a shared Google calendar (it should be visible on the trusted-substrate.org page).
>> >> >> >> I'll try to summarize what I learnt and highlight my view onwhat can be next steps in a future mail.
>> >> >> >> Cheers >> >> >> >> FF >> >> >> >> On Thu, 8 Apr 2021 at 13:56, Manish Pandey2 via TF-A <tf-a@lists.trustedfirmware.org> wrote:
>> >> Hi, >> >> >> >> From TF-A project point of view, we prefer to use existingmechanism to pass parameters across boot stages using linked list of tagged elements (as suggested by Julius). It has support for both generic and SiP-specific tags. Having said that, it does not stop partners to introduce new mechanisms suitable for their usecase in platform port initially and later move to generic code if its suitable for other platforms.
>> >> >> >> To start with, Ampere can introduce a platform specificimplementation of memory tag(speed/NUMA topology etc) which can be evaluated and discussed for generalization in future. The tag will be populated in BL2 stage and can be forwarded to further stages(and to BL33) by passing the head of list pointer in one of the registers. Initially any register can be used but going forward a standardization will be needed.
>> >> >> >> The U-boot bloblist mentioned by Simon is conceptually similarto what TF-A is using, if there is consensus of using bloblist/taglist then TF-A tag list may be enhanced to take best of both the implementations.
>> >> >> >> One of the potential problems of having structure used indifferent projects is maintainability, this can be avoided by having a single copy of these structures in TF-A (kept inside "include/export" which intended to be used by other projects.)
>> >> >> >> Regarding usage of either UUID or tag, I echo the sentiments ofSimon and Julius to keep it simple and use tag values.
>> >> >> >> Looking forward to having further discussions on zoom call today. >> >> >> >> Thanks >> >> Manish P >> >> >> >> ________________________________ >> >> From: TF-A <tf-a-bounces@lists.trustedfirmware.org> on behalfof Julius Werner via TF-A tf-a@lists.trustedfirmware.org
>> Sent: 25 March 2021 02:43 >> To: Simon Glass <sjg@chromium.org> >> Cc: Harb Abdulhamid OS <abdulhamid@os.amperecomputing.com>;Boot Architecture Mailman List boot-architecture@lists.linaro.org; tf-a@lists.trustedfirmware.org tf-a@lists.trustedfirmware.org; U-Boot Mailing List u-boot@lists.denx.de; Paul Isaac's paul.isaacs@linaro.org; Ron Minnich rminnich@google.com
>> Subject: Re: [TF-A] Proposal: TF-A to adopt hand-off blocks(HOBs) for information passing between boot stages
>> >> >> >> Just want to point out that TF-A currently already supports a(very simple) mechanism like this:
>> >> >> >>https://review.trustedfirmware.org/plugins/gitiles/TF-A/trusted-firmware-a/+...
>> >>https://review.trustedfirmware.org/plugins/gitiles/TF-A/trusted-firmware-a/+...
>> >>https://review.trustedfirmware.org/plugins/gitiles/TF-A/trusted-firmware-a/+...
>> >> >> >> It's just a linked list of tagged elements. The tag space issplit into TF-A-wide generic tags and SiP-specific tags (with plenty of room to spare if more areas need to be defined -- a 64-bit tag can fit a lot). This is currently being used by some platforms that run coreboot in place of BL1/BL2, to pass information from coreboot (BL2) to BL31.
>> >> >> >> I would echo Simon's sentiment of keeping this as simple aspossible and avoiding complicated and bloated data structures with UUIDs. You usually want to parse something like this as early as possible in the passed-to firmware stage, particularly if the structure encodes information about the debug console (like it does for the platforms I mentioned above). For example, in BL31 this basically means doing it right after moving from assembly to C in bl31_early_platform_setup2() to get the console up before running anything else. At that point in the BL31 initialization, the MMU and caches are disabled, so data accesses are pretty expensive and you don't want to spend a lot of parsing effort or calculate complicated checksums or the like. You just want something extremely simple where you ideally have to touch every data word only once.
>> >> >> >> On Wed, Mar 24, 2021 at 5:06 PM Simon Glass via TF-A <tf-a@lists.trustedfirmware.org> wrote:
>> >> Hi Harb, >> >> >> >> On Wed, 24 Mar 2021 at 11:39, Harb Abdulhamid OS <abdulhamid@os.amperecomputing.com> wrote:
>> >> Hello Folks, >> >> Appreciate the feedback and replies on this. Glad to see thatthere is interest in this topic.
>> >> >> >> I try to address the comments/feedback from Francois and Simonbelow….
>> >> >> >> @François Ozog – happy to discuss this on a zoom call. I willmake that time slot work, and will be available to attend April 8, 4pm CT.
>> >> >> >> Note that I’m using the term “HOB” here more generically, asthere are typically vendor specific structures beyond the resource descriptor HOB, which provides only a small subset of the information that needs to be passed between the boot phases.
>> >> >> >> The whole point here is to provide mechanism to develop firmwarethat we can build ARM Server SoC’s that support *any* BL33 payload (e.g. EDK2, AptioV, CoreBoot, and maybe even directly boot strapping LinuxBoot at some point). In other-words, we are trying to come up with a TF-A that would be completely agnostic to the implementation of BL33 (i.e. BL33 is built completely independently by a separate entity – e.g. an ODM/OEM).
>> >> >> >> Keep in mind, in the server/datacenter market segment we are notbuilding vertically integrated systems with a single entity compiling firmware/software stacks like most folks in TF-A have become use to. There are two categories of higher level firmware code blobs in the server/datacenter model:
>> >> “SoC” or “silicon” firmware – in TF-A this may map to BL1, BL2,BL31, and *possibly* one or more BL32 instances
>> “Platform” or “board” firmware – in TF-A this may map to BL33and *possibly* one or more BL32 instances.
>> >> >> >> Even the platform firmware stack could be further fragmented byhaving multiple entities involved in delivering the entire firmware stack: IBVs, ODMs, OEMs, CSPs, and possibly even device vendor code.
>> >> >> >> To support a broad range of platform designs with a broad rangeof memory devices, we need a crisp and clear contract between the SoC firmware that initializes memory (e.g. BL2) and how that platform boot firmware (e.g. BL33) gathers information about what memory that was initialized, at what speeds, NUMA topology, and many other relevant information that needs to be known and comprehended by the platform firmware and eventually by the platform software.
>> >> >> >> I understand the versatility of DT, but I see two major problemswith DT:
>> >> DT requires more complicated parsing to get properties, and evenmore complex to dynamically set properties – this HOB structures may need to be generated in boot phases where DDR is not available, and therefore we will be extremely memory constrained.
>> DT is probably overkill for this purpose – We really just want alist of pointers to simple C structures that code cast (e.g. JEDEC SPD data blob)
>> >> >> >> I think that we should not mix the efforts around DT/ACPI specswith what we are doing here, because those specs and concepts were developed for a completely different purpose (i.e. abstractions needed for OS / RTOS software, and not necessarily suitable for firmware-to-firmware hand-offs).
>> >> >> >> Frankly, I would personally push back pretty hard on definingSMC’s for something that should be one way information passing. Every SMC we add is another attack vector to the secure world and an increased burden on the folks that have to do security auditing and threat analysis. I see no benefit in exposing these boot/HOB/BOB structures at run-time via SMC calls.
>> >> >> >> Please do let me know if you disagree and why. Look forward todiscussing on this thread or on the call.
>> >> >> >> @Simon Glass - Thanks for the pointer to bloblist. I brieflyreviewed and it seems like a good baseline for what we may be looking for.
>> >> >> >> That being said, I would say that there is some benefit inhaving some kind of unique identifiers (e.g. UUID or some unique signature) so that we can tie standardized data structures (based on some future TBD specs) to a particular ID. For example, if the TPM driver in BL33 is looking for the TPM structure in the HOB/BOB list, and may not care about the other data blobs. The driver needs a way to identify and locate the blob it cares about.
>> >> >> >> The tag is intended to serve that purpose, although perhaps itshould switch from an auto-allocating enum to one with explicit values for each entry and a range for 'local' use.
>> >> >> >> I guess we can achieve this with the tag, but the problem withtag when you have eco-system with a lot of parties doing parallel development, you can end up with tag collisions and folks fighting about who has rights to what tag values. We would need some official process for folks to register tags for whatever new structures we define, or maybe some tag range for vendor specific structures. This comes with a lot of pain and bureaucracy. On the other hand, UUID has been a proven way to make it easy to just define your own blobs with *either* standard or vendor specific structures without worry of ID collisions between vendors.
>> >> >> >> True. I think the pain is overstated, though. In this case Ithink we actually want something that can be shared between projects and orgs, so some amount of coordination could be considered a benefit. It could just be a github pull request. I find the UUID unfriendly and not just to code size and eyesight! Trying to discover what GUIDs mean or are valid is quite tricky. E.g. see this code:
>> >> >> >> #define FSP_HOB_RESOURCE_OWNER_TSEG_GUID \ >> EFI_GUID(0xd038747c, 0xd00c, 0x4980, \ >> 0xb3, 0x19, 0x49, 0x01, 0x99, 0xa4, 0x7d, 0x55) >> >> (etc.) >> >> >> >> static struct guid_name { >> efi_guid_t guid; >> const char *name; >> } guid_name[] = { >> { FSP_HOB_RESOURCE_OWNER_TSEG_GUID, "TSEG" }, >> { FSP_HOB_RESOURCE_OWNER_FSP_GUID, "FSP" }, >> { FSP_HOB_RESOURCE_OWNER_SMM_PEI_SMRAM_GUID, "SMM PEI SMRAM"},
>> { FSP_NON_VOLATILE_STORAGE_HOB_GUID, "NVS" }, >> { FSP_VARIABLE_NV_DATA_HOB_GUID, "Variable NVS" }, >> { FSP_GRAPHICS_INFO_HOB_GUID, "Graphics info" }, >> { FSP_HOB_RESOURCE_OWNER_PCD_DATABASE_GUID1, "PCD databaseea" },
>> { FSP_HOB_RESOURCE_OWNER_PCD_DATABASE_GUID2, "PCD database9b" },
>> >> (never figured out what those two are) >> >> >> { FSP_HOB_RESOURCE_OWNER_PEIM_DXE_GUID, "PEIM Init DXE" }, >> { FSP_HOB_RESOURCE_OWNER_ALLOC_STACK_GUID, "Alloc stack" }, >> { FSP_HOB_RESOURCE_OWNER_SMBIOS_MEMORY_GUID, "SMBIOS memory"},
>> { {}, "zero-guid" }, >> {} >> }; >> >> static const char *guid_to_name(const efi_guid_t *guid) >> { >> struct guid_name *entry; >> >> for (entry = guid_name; entry->name; entry++) { >> if (!guidcmp(guid, &entry->guid)) >> return entry->name; >> } >> >> return NULL; >> } >> >> >> >> Believe it or not it took a fair bit of effort to find just thatsmall list, with nearly every one in a separate doc, from memory.
>> >> >> >> >> >> We can probably debate whether there is any value in GUID/UUIDor not during the call… but again, boblist seems like a reasonable starting point as an alternative to HOB.
>> >> >> >> Indeed. There is certainly value in both approaches. >> >> >> >> Regards, >> >> Simon >> >> >> >> >> >> Thanks, >> >> --Harb >> >> >> >> From: François Ozog <francois.ozog@linaro.org> >> Sent: Tuesday, March 23, 2021 10:00 AM >> To: François Ozog <francois.ozog@linaro.org>; Ron Minnich <rminnich@google.com>; Paul Isaac's paul.isaacs@linaro.org
>> Cc: Simon Glass <sjg@chromium.org>; Harb Abdulhamid OS <abdulhamid@os.amperecomputing.com>; Boot Architecture Mailman List < boot-architecture@lists.linaro.org>; tf-a@lists.trustedfirmware.org
>> Subject: Re: [TF-A] Proposal: TF-A to adopt hand-off blocks(HOBs) for information passing between boot stages
>> >> >> >> +Ron Minnich +Paul Isaac's >> >> >> >> Adding Ron and Paul because I think this interface should bealso benefiting LinuxBoot efforts.
>> >> >> >> On Tue, 23 Mar 2021 at 11:17, François Ozog via TF-A <tf-a@lists.trustedfirmware.org> wrote:
>> >> Hi, >> >> >> >> I propose we cover the topic at the next Trusted Substratezoom call on April 8th 4pm CET.
>> >> >> >> The agenda: >> >> ABI between non-secure firmware and the rest of firmware (EL3,S-EL1, S-EL2, SCP) to adapt hardware description to some runtime conditions.
>> >> runtime conditions here relates to DRAM size and topologydetection, secure DRAM memory carvings, PSCI and SCMI interface publishing.
>> >> >> >> For additional background on existing metadata: UEFI PlatformInitialization Specification Version 1.7, 5.5 Resource Descriptor HOB
>> >> Out of the ResourceType we care about isEFI_RESOURCE_SYSTEM_MEMORY.
>> >> This HOB lacks memory NUMA attachment or something that could berelated to fill SRAT table for ACPI or relevant DT proximity domains.
>> >> HOB is not consistent accros platforms: some platforms (Arm)lists memory from the booting NUMA node, other platforms (x86) lists all memory from all NUMA nodes. (At least this is the case on the two platforms I tested).
>> >> >> >> There are two proposals to use memory structures from SPL/BLx upto the handover function (as defined in the Device Tree technical report) which can be U-boot (BL33 or just U-Boot in case of SPL/U-Boot scheme) or EDK2.
>> >> I would propose we also discuss possibility of FF-A interface toactually query information or request actions to be done (this is a model actually used in some SoCs with proprietary SMC calls).
>> >> >> >> Requirements (to be validated): >> >> - ACPI and DT hardware descriptions. >> >> - agnostic to boot framework (SPL/U-Boot, TF-A/U-Boot, TF-A/EDK2) >> >> - agnostic to boot framework (SPL/U-Boot, TF-A/U-Boot,TF-A/EDK2, TF-A/LinuxBoot)
>> >> - at least allows complete DRAM description and "persistent"usage (reserved areas for secure world or other usages)
>> >> - support secure world device assignment >> >> >> >> Cheers >> >> >> >> FF >> >> >> >> >> >> On Mon, 22 Mar 2021 at 19:56, Simon Glass <sjg@chromium.org>wrote:
>> >> Hi, >> >> Can I suggest using bloblist for this instead? It is lightweight, >> easier to parse, doesn't have GUIDs and is already used withinU-Boot
>> for passing info between SPL/U-Boot, etc. >> >> Docs here:https://github.com/u-boot/u-boot/blob/master/doc/README.bloblist
>> Header file describes the format: >> https://github.com/u-boot/u-boot/blob/master/include/bloblist.h >> >> Full set of unit tests: >> https://github.com/u-boot/u-boot/blob/master/test/bloblist.c >> >> Regards, >> Simon >> >> On Mon, 22 Mar 2021 at 23:58, François Ozog <francois.ozog@linaro.org> wrote:
>> > >> > +Boot Architecture Mailman List <boot-architecture@lists.linaro.org>
>> > >> > standardization is very much welcomed here and need toaccommodate a very
>> > diverse set of situations. >> > For example, TEE OS may need to pass memory reservations toBL33 or
>> > "capture" a device for the secure world. >> > >> > I have observed a number of architectures: >> > 1) pass information from BLx to BLy in the form of a specificobject
>> > 2) BLx called by BLy by a platform specific SMC to getinformation
>> > 3) BLx called by BLy by a platform specific SMC to performDevice Tree
>> > fixups >> > >> > I also imagined a standardized "broadcast" FF-A call so thatany firmware
>> > element can either provide information or "do something". >> > >> > My understanding of your proposal is about standardizing onarchitecture 1)
>> > with the HOB format. >> > >> > The advantage of the HOB is simplicity but it may be difficultto implement
>> > schemes such as pruning a DT because device assignment in thesecure world.
>> > >> > In any case, it looks feasible to have TF-A and OP-TEEcomplement the list
>> > of HOBs to pass information downstream (the bootflow). >> > >> > It would be good to start with building the comprehensive listof
>> > information that need to be conveyed between firmware elements: >> > >> > information. | authoritative entity | reporting entity |information
>> > exchanged: >> > dram | TFA | TFA|>> > <format to be detailed, NUMA topology to build the SRAT tableor DT
>> > equivalent?> >> > PSCI | SCP | TFA?|>> > SCMI | SCP or TEE-OS | TFA? TEE-OS?| >> > secure SRAM | TFA. | TFA.|
>> > secure DRAM | TFA? TEE-OS? | TFA? TEE-OS? | >> > other? | | >> > | >> > >> > Cheers >> > >> > FF >> > >> > >> > On Mon, 22 Mar 2021 at 09:34, Harb Abdulhamid OS via TF-A < >> > tf-a@lists.trustedfirmware.org> wrote: >> > >> > > Hello Folks, >> > > >> > > >> > > >> > > I'm emailing to start an open discussion about the adoptionof a concept
>> > > known as "hand-off blocks" or HOB to become a part of theTF-A Firmware
>> > > Framework Architecture (FFA). This is something that is apretty major
>> > > pain point when it comes to the adoption of TF-A in ARMServer SoC’s
>> > > designed to enable a broad range of highly configurabledatacenter
>> > > platforms. >> > > >> > > >> > > >> > > >> > > >> > > What is a HOB (Background)? >> > > >> > > --------------------------- >> > > >> > > UEFI PI spec describes a particular definition for how HOBmay be used for
>> > > transitioning between the PEI and DXE boot phases, which isa good
>> > > reference point for this discussion, but not necessarily theexact solution
>> > > appropriate for TF-A. >> > > >> > > >> > > >> > > A HOB is simply a dynamically generated data structurepassed in between
>> > > two boot phases. This is information that was obtainedthrough discovery
>> > > and needs to be passed forward to the next boot phase*once*, with no API
>> > > needed to call back (e.g. no call back into previousfirmware phase is
>> > > needed to fetch this information at run-time - it is simplypassed one time
>> > > during boot). >> > > >> > > >> > > >> > > There may be one or more HOBs passed in between bootphases. If there are
>> > > more than one HOB that needs to be passed, this can be in aform of a "HOB
>> > > table", which (for example) could be a UUID indexed array ofpointers to
>> > > HOB structures, used to locate a HOB of interest (based onUUID). In such
>> > > cases, instead of passing a single HOB, the boot phases mayrely on passing
>> > > the pointer to the HOB table. >> > > >> > > >> > > >> > > This has been extremely useful concept to employ on highlyconfigurable
>> > > systems that must rely on flexible discovery mechanisms toinitialize and
>> > > boot the system. This is especially helpful when you havemultiple
>> > > >> > > >> > > >> > > >> > > >> > > Why do we need HOBs in TF-A?: >> > > >> > > ----------------------------- >> > > >> > > It is desirable that EL3 firmware (e.g. TF-A) built for ARMServer SoC in
>> > > a way that is SoC specific *but* platform agnostic. Thismeans that a
>> > > single ARM SoC that a SiP may deliver to customers mayprovide a single
>> > > TF-A binary (e.g. BL1, BL2, BL31) that could be used tosupport a broad
>> > > range of platform designs and configurations in order toboot a platform
>> > > specific firmware (e.g. BL33 and possibly even BL32 code).In order to
>> > > achieve this, the platform configuration must be*discovered* instead of
>> > > statically compiled as it is today in TF-A via device treebased
>> > > enumeration. The mechanisms of discovery may differ broadlydepending on
>> > > the relevant industry standard, or in some cases may haverely on SiP
>> > > specific discovery flows. >> > > >> > > >> > > >> > > For example: On server systems that support a broad rangeDIMM memory
>> > > population/topologies, all the necessary informationrequired to boot is
>> > > fully discovered via standard JEDEC Serial Presence Detect(SPD) over an
>> > > I2C bus. Leveraging the SPD bus, may platform variantscould be supported
>> > > with a single TF-A binary. Not only is this informationrequired to
>> > > initialize memory in early boot phases (e.g. BL2), thesubsequent boot
>> > > phases will also need this SPD info to construct a systemphysical address
>> > > map and properly initialize the MMU based on the memorypresent, and where
>> > > the memory may be present. Subsequent boot phases (e.g.BL33 / UEFI) may
>> > > need to generate standard firmware tables to the operatingsystems, such as
>> > > SMBIOS tables describing DIMM topology and various ACPItables (e.g. SLIT,
>> > > SRAT, even NFIT if NVDIMM's are present). >> > > >> > > >> > > >> > > In short, it all starts with a standardized or vendorspecific discovery
>> > > flow in an early boot stage (e.g. BL1/BL2), followed by thepassing of
>> > > information to the next boot stages (e.g. BL31/BL32/BL33). >> > > >> > > >> > > >> > > Today, every HOB may be a vendor specific structure, but inthe future
>> > > there may be benefit of defining standard HOBs. This may beuseful for
>> > > memory discovery, passing the system physical address map,enabling TPM
>> > > measured boot, and potentially many other common HOBuse-cases.
>> > > >> > > >> > > >> > > It would be extremely beneficial to the datacenter marketsegment if the
>> > > TF-A community would adopt this concept of informationpassing between all
>> > > boot phases as opposed to rely solely on device treeenumeration. This is
>> > > not intended to replace device tree, rather intended as analternative way
>> > > to describe the info that must be discovered and dynamicallygenerated.
>> > > >> > > >> > > >> > > >> > > >> > > Conclusion: >> > > >> > > ----------- >> > > >> > > We are proposing that the TF-A community begin pursuing theadoption of
>> > > HOBs as a mechanism used for information exchange betweeneach boot stage
>> > > (e.g. BL1->BL2, BL2->BL31, BL31->BL32, and BL31->BL33)?Longer term we
>> > > want to explore standardizing some HOB structures for theBL33 phase (e.g.
>> > > UEFI HOB structures), but initially would like to agree onthis being a
>> > > useful mechanism used to pass information between each bootstage.
>> > > >> > > >> > > >> > > Thanks, >> > > >> > > --Harb >> > > >> > > >> > > >> > > >> > > >> > > >> > > -- >> > > TF-A mailing list >> > > TF-A@lists.trustedfirmware.org >> > > https://lists.trustedfirmware.org/mailman/listinfo/tf-a >> > > >> > >> > >> > -- >> > François-Frédéric Ozog | *Director Linaro Edge & Fog ComputingGroup*
>> > T: +33.67221.6485 >> > francois.ozog@linaro.org | Skype: ffozog >> > _______________________________________________ >> > boot-architecture mailing list >> > boot-architecture@lists.linaro.org >> > https://lists.linaro.org/mailman/listinfo/boot-architecture >> >> >> >> >> -- >> >> François-Frédéric Ozog | Director Linaro Edge & Fog ComputingGroup
>> >> T: +33.67221.6485 >> francois.ozog@linaro.org | Skype: ffozog >> >> >> >> -- >> TF-A mailing list >> TF-A@lists.trustedfirmware.org >> https://lists.trustedfirmware.org/mailman/listinfo/tf-a >> >> >> >> >> -- >> >> François-Frédéric Ozog | Director Linaro Edge & Fog ComputingGroup
>> >> T: +33.67221.6485 >> francois.ozog@linaro.org | Skype: ffozog >> >> >> >> -- >> TF-A mailing list >> TF-A@lists.trustedfirmware.org >> https://lists.trustedfirmware.org/mailman/listinfo/tf-a >> >> -- >> TF-A mailing list >> TF-A@lists.trustedfirmware.org >> https://lists.trustedfirmware.org/mailman/listinfo/tf-a >> >> >> >> >> -- >> >> François-Frédéric Ozog | Director Linaro Edge & Fog ComputingGroup
>> >> T: +33.67221.6485 >> francois.ozog@linaro.org | Skype: ffozog >> >> > > -- > TF-A mailing list > TF-A@lists.trustedfirmware.org > https://lists.trustedfirmware.org/mailman/listinfo/tf-a -- TF-A mailing list TF-A@lists.trustedfirmware.org https://lists.trustedfirmware.org/mailman/listinfo/tf-a-- TF-A mailing list TF-A@lists.trustedfirmware.org https://lists.trustedfirmware.org/mailman/listinfo/tf-a
participants (1)
-
François Ozog