- Devel - lists.tockos.org

Re: Call for Tock meeting agenda items

by Amit Levy

- Working Group reorganization: https://lists.tockos.org/hyperkitty/list/devel@lists.tockos.org/thread/343B… - Please read in advance if you haven't already - I estimate 10 minutes, max - Progress on Core WG assigned PRs - I will generated a summary for this first time before the meeting - Could replace/be part of updates. 5 minutes max. - If there are particular discussion-worthy issues with assigned PRs, please add as agenda item - To revive or not to revive updating the storage HILs (two stale PRs): - https://github.com/tock/tock/pull/2248 - https://github.com/tock/tock/pull/2993 Brad Campbell <bradjc(a)virginia.edu> writes: > - Kernel testing +1 > - Static mut and updating nightly +1 > On Thu, Feb 22, 2024 at 8:46 PM Branden Ghena <branden(a)northwestern.edu> > wrote: > >> It would be good to at least discuss moving forward on >> https://github.com/tock/tock/pull/3822 >> * Do we want review assignments? (Yes from me) >> * What are the minimal features for it to start with? (we could add more >> later) >> >> >> On Thu, Feb 22, 2024 at 1:05 PM Brad Campbell <bradjc(a)virginia.edu> wrote: >> >>> Topics I would like to discuss for this week: >>> >>> - Digest HIL updates https://github.com/tock/tock/pull/3866 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3866__;!!Dq0X…> >>> - Both digests and digest implementations are now strongly >>> templated on a particular algorithm, rather than just the length of the >>> resulting digest. >>> - This is motivated by being able to strongly type cryptographic >>> signatures to a particular digest type. >>> - Kernel testing >>> - RFC: "Configuration" boards. These would be "mini" boards with >>> enough functionality to test different configurations of the Tock kernel. >>> https://github.com/tock/tock/pull/3873 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3873__;!!Dq0X…> >>> - Trait to wrap capsule tests: >>> https://github.com/tock/tock/pull/3871 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3871__;!!Dq0X…>. >>> This allows the test runner to learn when a test has finished, since the >>> test might run many async calls. >>> - Static mut and updating nightly >>> - We currently cannot compile the kernel on the latest nightly: >>> https://github.com/tock/tock/issues/3841 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/issues/3841__;!!Dq…> >>> - We don't want to get stuck on an old version of rust, and it >>> seems like rust wants us to tackle this problem now. >>> - Is anyone willing to develop a "SyncCell" (or whatever the >>> correct name is)? I propose we put a time bound on this, and use the >>> rust-suggested "fix" if we don't have anything better so we can update >>> nightly. >>> - PRs >>> - https://github.com/tock/tock/pull/3855 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3855__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3859 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3859__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3843 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3843__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3869 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3869__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3860 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3860__;!!Dq0X…> >>> >>> >>> On Fri, Feb 16, 2024 at 9:28 AM Brad Campbell <bradjc(a)virginia.edu> >>> wrote: >>> >>>> Unfortunately I have to leave at half past today. >>>> >>>> >>>> - https://github.com/tock/tock/pull/3849 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3849__;!!Dq0X…> >>>> - Async process loader >>>> - How to get PRs commented on or merged? >>>> - https://github.com/tock/tock/pull/3817 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3817__;!!Dq0X…> >>>> - https://github.com/tock/tock/pull/3836 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3836__;!!Dq0X…> >>>> - https://github.com/tock/tock/pull/3843 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3843__;!!Dq0X…> >>>> - https://github.com/tock/tock/pull/3846 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3846__;!!Dq0X…> >>>> - Looking for ideas here...do we need a new review process? All >>>> core members are supposed to comment within 7 days. >>>> - My forever quest to start supporting signed processes: >>>> https://github.com/tock/tock/pull/3772 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>>> :( >>>> - libtock-c refactor proposal >>>> - >>>> https://github.com/tock/libtock-c/blob/libtock-folders/doc/guide.md >>>> <https://urldefense.com/v3/__https://github.com/tock/libtock-c/blob/libtock-…> >>>> - libtock-c >>>> - Don't compile extern libraries with -Werror: >>>> https://github.com/tock/libtock-c/pull/368 >>>> <https://urldefense.com/v3/__https://github.com/tock/libtock-c/pull/368__;!!…> >>>> >>>> >>>> On Fri, Feb 9, 2024 at 9:21 AM Brad Campbell <bradjc(a)virginia.edu> >>>> wrote: >>>> >>>>> Ideas for today: >>>>> >>>>> >>>>> - Stable Rust - https://github.com/tock/tock/pull/3802 >>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3802__;!!Dq0X…> >>>>> - Signature HIL - https://github.com/tock/tock/pull/3772 >>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>>>> - libtock-c refresh - How to make it more user friendly? >>>>> - What types of changes would make libtock-c easier to use? >>>>> Where are the pain points? >>>>> - Do we need yield-wait-for-x? >>>>> - Should we copy some existing interface? >>>>> - We probably aren't going to be able to do an entire rewrite, >>>>> but are there manageable steps that would make a difference? >>>>> >>>>> >>>>> On Fri, Feb 2, 2024 at 11:24 AM Brad Campbell <bradjc(a)virginia.edu> >>>>> wrote: >>>>> >>>>>> >>>>>> - Process Checking Infrastructure PRs >>>>>> - https://github.com/tock/tock/pull/3772 >>>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>>>>> - https://github.com/tock/tock/pull/3818 >>>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3818__;!!Dq0X…> >>>>>> - Remove a nightly feature >>>>>> - https://github.com/tock/tock/pull/3803 >>>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3803__;!!Dq0X…> >>>>>> - Finalize Legal TRD >>>>>> - https://github.com/tock/tock/pull/3821 >>>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3821__;!!Dq0X…> >>>>>> >>>>>> >>>>>> On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: >>>>>> >>>>>>> Please reply with agenda items and +1s for today's meeting. >>>>>>> >>>>>>> Hudson is away today, so I'll be leading, hence the later than usual >>>>>>> request for agenda items. Luckily I think there are plenty. I suggest at >>>>>>> least: >>>>>>> >>>>>>> - Brad's suggestion for a new documentation working group (there is >>>>>>> already approval for this, but we can maybe discuss whether/how to support >>>>>>> this kind of process more generally. could be very brief) >>>>>>> _______________________________________________ >>>>>>> Devel mailing list -- devel(a)lists.tockos.org >>>>>>> To unsubscribe send an email to devel-leave(a)lists.tockos.org >>>>>>> >>>>>> _______________________________________________ >> Devel mailing list -- devel(a)lists.tockos.org >> To unsubscribe send an email to devel-leave(a)lists.tockos.org >> > _______________________________________________ > Devel mailing list -- devel(a)lists.tockos.org > To unsubscribe send an email to devel-leave(a)lists.tockos.org

Feb. 29, 2024

Re: Call for Tock meeting agenda items

by Brad Campbell

- Kernel testing - RFC: "Configuration" boards. These would be "mini" boards with enough functionality to test different configurations of the Tock kernel. https://github.com/tock/tock/pull/3873 - Trait to wrap capsule tests: https://github.com/tock/tock/pull/3871. This allows the test runner to learn when a test has finished, since the test might run many async calls. - Static mut and updating nightly - We currently cannot compile the kernel on the latest nightly: https://github.com/tock/tock/issues/3841 <https://github.com/tock/tock/issues/3841> - We don't want to get stuck on an old version of rust, and it seems like rust wants us to tackle this problem now. - Is anyone willing to develop a "SyncCell" (or whatever the correct name is)? I propose we put a time bound on this, and use the rust-suggested "fix" if we don't have anything better so we can update nightly. On Thu, Feb 22, 2024 at 8:46 PM Branden Ghena <branden(a)northwestern.edu> wrote: > It would be good to at least discuss moving forward on > https://github.com/tock/tock/pull/3822 > * Do we want review assignments? (Yes from me) > * What are the minimal features for it to start with? (we could add more > later) > > > On Thu, Feb 22, 2024 at 1:05 PM Brad Campbell <bradjc(a)virginia.edu> wrote: > >> Topics I would like to discuss for this week: >> >> - Digest HIL updates https://github.com/tock/tock/pull/3866 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3866__;!!Dq0X…> >> - Both digests and digest implementations are now strongly >> templated on a particular algorithm, rather than just the length of the >> resulting digest. >> - This is motivated by being able to strongly type cryptographic >> signatures to a particular digest type. >> - Kernel testing >> - RFC: "Configuration" boards. These would be "mini" boards with >> enough functionality to test different configurations of the Tock kernel. >> https://github.com/tock/tock/pull/3873 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3873__;!!Dq0X…> >> - Trait to wrap capsule tests: >> https://github.com/tock/tock/pull/3871 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3871__;!!Dq0X…>. >> This allows the test runner to learn when a test has finished, since the >> test might run many async calls. >> - Static mut and updating nightly >> - We currently cannot compile the kernel on the latest nightly: >> https://github.com/tock/tock/issues/3841 >> <https://urldefense.com/v3/__https://github.com/tock/tock/issues/3841__;!!Dq…> >> - We don't want to get stuck on an old version of rust, and it >> seems like rust wants us to tackle this problem now. >> - Is anyone willing to develop a "SyncCell" (or whatever the >> correct name is)? I propose we put a time bound on this, and use the >> rust-suggested "fix" if we don't have anything better so we can update >> nightly. >> - PRs >> - https://github.com/tock/tock/pull/3855 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3855__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3859 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3859__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3843 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3843__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3869 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3869__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3860 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3860__;!!Dq0X…> >> >> >> On Fri, Feb 16, 2024 at 9:28 AM Brad Campbell <bradjc(a)virginia.edu> >> wrote: >> >>> Unfortunately I have to leave at half past today. >>> >>> >>> - https://github.com/tock/tock/pull/3849 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3849__;!!Dq0X…> >>> - Async process loader >>> - How to get PRs commented on or merged? >>> - https://github.com/tock/tock/pull/3817 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3817__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3836 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3836__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3843 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3843__;!!Dq0X…> >>> - https://github.com/tock/tock/pull/3846 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3846__;!!Dq0X…> >>> - Looking for ideas here...do we need a new review process? All >>> core members are supposed to comment within 7 days. >>> - My forever quest to start supporting signed processes: >>> https://github.com/tock/tock/pull/3772 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>> :( >>> - libtock-c refactor proposal >>> - >>> https://github.com/tock/libtock-c/blob/libtock-folders/doc/guide.md >>> <https://urldefense.com/v3/__https://github.com/tock/libtock-c/blob/libtock-…> >>> - libtock-c >>> - Don't compile extern libraries with -Werror: >>> https://github.com/tock/libtock-c/pull/368 >>> <https://urldefense.com/v3/__https://github.com/tock/libtock-c/pull/368__;!!…> >>> >>> >>> On Fri, Feb 9, 2024 at 9:21 AM Brad Campbell <bradjc(a)virginia.edu> >>> wrote: >>> >>>> Ideas for today: >>>> >>>> >>>> - Stable Rust - https://github.com/tock/tock/pull/3802 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3802__;!!Dq0X…> >>>> - Signature HIL - https://github.com/tock/tock/pull/3772 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>>> - libtock-c refresh - How to make it more user friendly? >>>> - What types of changes would make libtock-c easier to use? >>>> Where are the pain points? >>>> - Do we need yield-wait-for-x? >>>> - Should we copy some existing interface? >>>> - We probably aren't going to be able to do an entire rewrite, >>>> but are there manageable steps that would make a difference? >>>> >>>> >>>> On Fri, Feb 2, 2024 at 11:24 AM Brad Campbell <bradjc(a)virginia.edu> >>>> wrote: >>>> >>>>> >>>>> - Process Checking Infrastructure PRs >>>>> - https://github.com/tock/tock/pull/3772 >>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>>>> - https://github.com/tock/tock/pull/3818 >>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3818__;!!Dq0X…> >>>>> - Remove a nightly feature >>>>> - https://github.com/tock/tock/pull/3803 >>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3803__;!!Dq0X…> >>>>> - Finalize Legal TRD >>>>> - https://github.com/tock/tock/pull/3821 >>>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3821__;!!Dq0X…> >>>>> >>>>> >>>>> On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: >>>>> >>>>>> Please reply with agenda items and +1s for today's meeting. >>>>>> >>>>>> Hudson is away today, so I'll be leading, hence the later than usual >>>>>> request for agenda items. Luckily I think there are plenty. I suggest at >>>>>> least: >>>>>> >>>>>> - Brad's suggestion for a new documentation working group (there is >>>>>> already approval for this, but we can maybe discuss whether/how to support >>>>>> this kind of process more generally. could be very brief) >>>>>> _______________________________________________ >>>>>> Devel mailing list -- devel(a)lists.tockos.org >>>>>> To unsubscribe send an email to devel-leave(a)lists.tockos.org >>>>>> >>>>> _______________________________________________ > Devel mailing list -- devel(a)lists.tockos.org > To unsubscribe send an email to devel-leave(a)lists.tockos.org >

Feb. 29, 2024

RFC: Revamping Working Groups

by Amit Levy

Hi all, Please read and give feedback on a high-level proposal to change the role of the Core WG below (starts after my sign-off). If you are a Core WG member, contributor to the Tock repo or other repositories owned by the Tock GitHub organization, have considered contributing to Tock, or otherwise have a stake in Tock's direction, this affects you. If none-of-the-above apply, you may have been mistakenly subscribed to this mailing list :) Note that the proposal is a draft with minimal feedback from some other Core WG members, but is /not/ an authoritative version in any sense. Ok... So: We have a code review and issue triage crisis*! The scope and pace of Tock development has outgrown the ability of a concentrated core maintainer group to keep up with while also shepherding important foundational aspects of Tock. To address this, I propose that we evolve how the Tock project is managed. The /vision/ is responsibility for Tock subsystems as well as related ecosystem tools/projects that is completely devolved to mostly independent working groups. The Core WG, in this vision, would primarily serve to coordinate between working groups, establish shared design vision, and establish/disband working groups as the needs of the project changes. Concrete working groups, on the other hand would take on PR review, design discussion, contribution guidelines, etc for their own purviews. Working groups also facilitate common spaces for contributors and users focused on specific areas to interact and collaborate. I think moving in this direction is important for supporting a growing community and for avoiding some past mistakes in sherpherding and encouraging upstream contribution, as well as making long-term maintanence more reasonable. Below is a draft proposal for a first step in this direction. The gist is to formally shift the purview of the Core WG from approving all PRs to empowering other WGs with code review and merging privileges and serving as a backstop for issues and PRs that either fall outside of WGs or cross WG boundaries. But I think it's a start. Along with this high level proposal for a role-change of the Core WG, I envision we'd start with a working groups associated with the following areas and purviews: - Core - Kernel - The kernel crate (minus things like HILs) - Communication (FKA networking) - capsules/src/extra/net - capsules/src/extra/ieee802154 - capsules/src/extra/ble* - capsules/src/extra/usb - capsules/src/extra/can[.rs] - RISC-V - chips/ibex - chips/lowrisc - boards/opentitan - boards/litex - arch/rv32i - Userspace support for RISC-V - Documentation - tock-book - doc/ - User-space libraries - libtock-c - libtock-rs - libtock-[whatever-is-next] - Testing and development infrastructure - tockloader - CI and test infra - `tools/` - Cryptography - Capsules and HILs for AES, digests, signatures... you know... crypto stuff - Peripherals - Most of the rest of capsules, but specifically peripheral busses (i2c, spi, etc..), sensors, etc I think actually getting there is going to require more work than just what is spelled out below: actually establishing some more working groups and recruiting people to join them, probably some organization restructuring of the main Tock source tree to better match the "right" division of responsibilities, more explicit and written down shared guidelines on what it are universal requirements for code upstream (what does "quality code" mean, what are minimum testing and documentation requirements, etc), and maybe others. But this is a first step. -Amit *Am I exaggerating? Yes. --- # Tock Working Groups Working groups are focused groups to organize development around a particular aspect of Tock. ## Motivation Tock relies on contributions from developers with across a variety of organizations, physical distance, timezones, and engagement. It has also grown large in scope. This has resulted in strain on maintainers to regularly keep up with contributions while ensuring quality and important global policies like safety and security. It has also resulted in contributors' work going unaddressed for long periods of time. Tock encompases a large and varied set of subsystems, architectures, focus areas, libraries, ancillary tools, and documentation. Most contributors have expertise and stake in a subset of these. Moreover, it is impractical for any maintainer to keep up with the discussions and direction of each part of the project. Finally, different parts of the project should be able to move at different paces, with different levels of scrutiny. For example, a soundness bug or performance regression in the kernel crate can catestrophically impact all users and should be avoided if at all possible, while a sub-optimal design decision in an experimental user-space library is not a big deal. To facilitate this, working groups take on responsibility for specific sub-areas of the project. Members of a working group become experts in that sub-area and are best able to determine appropriate scrutiny for accepting contributions, frequency and mode of design discussions, etc. ## Structure And Responsibilities Tock development organizes around a core working group as well as additional area-specific working groups. The core working group oversees the project wholistically, defining high-level design goals and project direction, establishing working groups, and facilitating work that spans multiple working groups. Other working groups facilitate contributions to specific sub-areas of the project, with devolved decision making resposibility for accepting contributions, design, and direction. While working groups *oversee* development, working group members are not expected to be the primary source of contributions. Instead, working groups establish code review standards, define and communicate specific design direction for their purview, and ensure relevant contributors are both supported and effective. ### Working Group Organizational Guidelines Each working group will have a Lead who assembles the working group membership and is responsible for its operation. Each working group should include at least one member from the Core Working Group to ensure that the Core Working Group is regularly updated on the activities and motivations of the working group. The Core Working Group member need not be the working group Lead. Each working group, including the Core Working Group, establishes its own rules and procedures for accepting contributions, communicating, meeting, and making decisions. In absense of such rules, working groups will make decisions by consensus, have weekly voice calls, make meeting notes available publicly, and communicate asynchronously via a mailing list. However, working groups are encouraged to avoid consensus-based decision making as quickly as possible and establish appropriate meeting frequency and communication mode for their needs and membership. ## Core Working Group The Core Working Group shepherds and oversees the Tock OS and related tools and libraries. Importantly, it serves as a backstop for managing contributions that fall outside the purview of existing working groups. It also establishes new working groups to handle such contributions, and dissolves or re-organizes existing working groups. Formally, the Core Working Group controls who can directly commit to all Tock project repositories and devolves the ability to commit to specific repositories or components of repositories to other working groups. The Core Working Group, like other working groups, establishes its own rules for deciding how to accept contributions as well as how to establish and disband working groups. The Core Working Group's duties are: - Managing and overseeing code, documentation, testing, and releases for the Tock project. - Defining and communicating overall project goals and direction. - Establishing and delegating responsility over components and sub-projects to working groups. - Ensuring that working groups have the people and resources needed to accomplish their work. - Ensuring working groups are accountable to their delegated responsibilities and the project as a whole. - Coordinating decisions including (but not limited to) code, documentation, testing, and releases that affect purviews delegated to more than one working group. - Facilitating communication channels and consensus among working groups. - Coordinating project-wide changes to teams, structures, or processes.

Feb. 28, 2024

Re: Call for Tock meeting agenda items

by Branden Ghena

It would be good to at least discuss moving forward on https://github.com/tock/tock/pull/3822 * Do we want review assignments? (Yes from me) * What are the minimal features for it to start with? (we could add more later) On Thu, Feb 22, 2024 at 1:05 PM Brad Campbell <bradjc(a)virginia.edu> wrote: > Topics I would like to discuss for this week: > > - Digest HIL updates https://github.com/tock/tock/pull/3866 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3866__;!!Dq0X…> > - Both digests and digest implementations are now strongly > templated on a particular algorithm, rather than just the length of the > resulting digest. > - This is motivated by being able to strongly type cryptographic > signatures to a particular digest type. > - Kernel testing > - RFC: "Configuration" boards. These would be "mini" boards with > enough functionality to test different configurations of the Tock kernel. > https://github.com/tock/tock/pull/3873 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3873__;!!Dq0X…> > - Trait to wrap capsule tests: > https://github.com/tock/tock/pull/3871 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3871__;!!Dq0X…>. > This allows the test runner to learn when a test has finished, since the > test might run many async calls. > - Static mut and updating nightly > - We currently cannot compile the kernel on the latest nightly: > https://github.com/tock/tock/issues/3841 > <https://urldefense.com/v3/__https://github.com/tock/tock/issues/3841__;!!Dq…> > - We don't want to get stuck on an old version of rust, and it > seems like rust wants us to tackle this problem now. > - Is anyone willing to develop a "SyncCell" (or whatever the > correct name is)? I propose we put a time bound on this, and use the > rust-suggested "fix" if we don't have anything better so we can update > nightly. > - PRs > - https://github.com/tock/tock/pull/3855 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3855__;!!Dq0X…> > - https://github.com/tock/tock/pull/3859 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3859__;!!Dq0X…> > - https://github.com/tock/tock/pull/3843 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3843__;!!Dq0X…> > - https://github.com/tock/tock/pull/3869 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3869__;!!Dq0X…> > - https://github.com/tock/tock/pull/3860 > <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3860__;!!Dq0X…> > > > On Fri, Feb 16, 2024 at 9:28 AM Brad Campbell <bradjc(a)virginia.edu> wrote: > >> Unfortunately I have to leave at half past today. >> >> >> - https://github.com/tock/tock/pull/3849 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3849__;!!Dq0X…> >> - Async process loader >> - How to get PRs commented on or merged? >> - https://github.com/tock/tock/pull/3817 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3817__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3836 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3836__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3843 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3843__;!!Dq0X…> >> - https://github.com/tock/tock/pull/3846 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3846__;!!Dq0X…> >> - Looking for ideas here...do we need a new review process? All >> core members are supposed to comment within 7 days. >> - My forever quest to start supporting signed processes: >> https://github.com/tock/tock/pull/3772 >> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >> :( >> - libtock-c refactor proposal >> - >> https://github.com/tock/libtock-c/blob/libtock-folders/doc/guide.md >> <https://urldefense.com/v3/__https://github.com/tock/libtock-c/blob/libtock-…> >> - libtock-c >> - Don't compile extern libraries with -Werror: >> https://github.com/tock/libtock-c/pull/368 >> <https://urldefense.com/v3/__https://github.com/tock/libtock-c/pull/368__;!!…> >> >> >> On Fri, Feb 9, 2024 at 9:21 AM Brad Campbell <bradjc(a)virginia.edu> wrote: >> >>> Ideas for today: >>> >>> >>> - Stable Rust - https://github.com/tock/tock/pull/3802 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3802__;!!Dq0X…> >>> - Signature HIL - https://github.com/tock/tock/pull/3772 >>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>> - libtock-c refresh - How to make it more user friendly? >>> - What types of changes would make libtock-c easier to use? Where >>> are the pain points? >>> - Do we need yield-wait-for-x? >>> - Should we copy some existing interface? >>> - We probably aren't going to be able to do an entire rewrite, >>> but are there manageable steps that would make a difference? >>> >>> >>> On Fri, Feb 2, 2024 at 11:24 AM Brad Campbell <bradjc(a)virginia.edu> >>> wrote: >>> >>>> >>>> - Process Checking Infrastructure PRs >>>> - https://github.com/tock/tock/pull/3772 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3772__;!!Dq0X…> >>>> - https://github.com/tock/tock/pull/3818 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3818__;!!Dq0X…> >>>> - Remove a nightly feature >>>> - https://github.com/tock/tock/pull/3803 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3803__;!!Dq0X…> >>>> - Finalize Legal TRD >>>> - https://github.com/tock/tock/pull/3821 >>>> <https://urldefense.com/v3/__https://github.com/tock/tock/pull/3821__;!!Dq0X…> >>>> >>>> >>>> On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: >>>> >>>>> Please reply with agenda items and +1s for today's meeting. >>>>> >>>>> Hudson is away today, so I'll be leading, hence the later than usual >>>>> request for agenda items. Luckily I think there are plenty. I suggest at >>>>> least: >>>>> >>>>> - Brad's suggestion for a new documentation working group (there is >>>>> already approval for this, but we can maybe discuss whether/how to support >>>>> this kind of process more generally. could be very brief) >>>>> _______________________________________________ >>>>> Devel mailing list -- devel(a)lists.tockos.org >>>>> To unsubscribe send an email to devel-leave(a)lists.tockos.org >>>>> >>>>

Feb. 23, 2024

Re: Call for Tock meeting agenda items

by Brad Campbell

Topics I would like to discuss for this week: - Digest HIL updates https://github.com/tock/tock/pull/3866 - Both digests and digest implementations are now strongly templated on a particular algorithm, rather than just the length of the resulting digest. - This is motivated by being able to strongly type cryptographic signatures to a particular digest type. - Kernel testing - RFC: "Configuration" boards. These would be "mini" boards with enough functionality to test different configurations of the Tock kernel. https://github.com/tock/tock/pull/3873 - Trait to wrap capsule tests: https://github.com/tock/tock/pull/3871. This allows the test runner to learn when a test has finished, since the test might run many async calls. - Static mut and updating nightly - We currently cannot compile the kernel on the latest nightly: https://github.com/tock/tock/issues/3841 - We don't want to get stuck on an old version of rust, and it seems like rust wants us to tackle this problem now. - Is anyone willing to develop a "SyncCell" (or whatever the correct name is)? I propose we put a time bound on this, and use the rust-suggested "fix" if we don't have anything better so we can update nightly. - PRs - https://github.com/tock/tock/pull/3855 - https://github.com/tock/tock/pull/3859 - https://github.com/tock/tock/pull/3843 - https://github.com/tock/tock/pull/3869 - https://github.com/tock/tock/pull/3860 On Fri, Feb 16, 2024 at 9:28 AM Brad Campbell <bradjc(a)virginia.edu> wrote: > Unfortunately I have to leave at half past today. > > > - https://github.com/tock/tock/pull/3849 - Async process loader > - How to get PRs commented on or merged? > - https://github.com/tock/tock/pull/3817 > - https://github.com/tock/tock/pull/3836 > - https://github.com/tock/tock/pull/3843 > - https://github.com/tock/tock/pull/3846 > - Looking for ideas here...do we need a new review process? All > core members are supposed to comment within 7 days. > - My forever quest to start supporting signed processes: > https://github.com/tock/tock/pull/3772 :( > - libtock-c refactor proposal > - > https://github.com/tock/libtock-c/blob/libtock-folders/doc/guide.md > - libtock-c > - Don't compile extern libraries with -Werror: > https://github.com/tock/libtock-c/pull/368 > > > On Fri, Feb 9, 2024 at 9:21 AM Brad Campbell <bradjc(a)virginia.edu> wrote: > >> Ideas for today: >> >> >> - Stable Rust - https://github.com/tock/tock/pull/3802 >> - Signature HIL - https://github.com/tock/tock/pull/3772 >> - libtock-c refresh - How to make it more user friendly? >> - What types of changes would make libtock-c easier to use? Where >> are the pain points? >> - Do we need yield-wait-for-x? >> - Should we copy some existing interface? >> - We probably aren't going to be able to do an entire rewrite, but >> are there manageable steps that would make a difference? >> >> >> On Fri, Feb 2, 2024 at 11:24 AM Brad Campbell <bradjc(a)virginia.edu> >> wrote: >> >>> >>> - Process Checking Infrastructure PRs >>> - https://github.com/tock/tock/pull/3772 >>> - https://github.com/tock/tock/pull/3818 >>> - Remove a nightly feature >>> - https://github.com/tock/tock/pull/3803 >>> - Finalize Legal TRD >>> - https://github.com/tock/tock/pull/3821 >>> >>> >>> On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: >>> >>>> Please reply with agenda items and +1s for today's meeting. >>>> >>>> Hudson is away today, so I'll be leading, hence the later than usual >>>> request for agenda items. Luckily I think there are plenty. I suggest at >>>> least: >>>> >>>> - Brad's suggestion for a new documentation working group (there is >>>> already approval for this, but we can maybe discuss whether/how to support >>>> this kind of process more generally. could be very brief) >>>> _______________________________________________ >>>> Devel mailing list -- devel(a)lists.tockos.org >>>> To unsubscribe send an email to devel-leave(a)lists.tockos.org >>>> >>>

Feb. 22, 2024

Re: Call for Tock meeting agenda items

by Brad Campbell

Unfortunately I have to leave at half past today. - https://github.com/tock/tock/pull/3849 - Async process loader - How to get PRs commented on or merged? - https://github.com/tock/tock/pull/3817 - https://github.com/tock/tock/pull/3836 - https://github.com/tock/tock/pull/3843 - https://github.com/tock/tock/pull/3846 - Looking for ideas here...do we need a new review process? All core members are supposed to comment within 7 days. - My forever quest to start supporting signed processes: https://github.com/tock/tock/pull/3772 :( - libtock-c refactor proposal - https://github.com/tock/libtock-c/blob/libtock-folders/doc/guide.md - libtock-c - Don't compile extern libraries with -Werror: https://github.com/tock/libtock-c/pull/368 On Fri, Feb 9, 2024 at 9:21 AM Brad Campbell <bradjc(a)virginia.edu> wrote: > Ideas for today: > > > - Stable Rust - https://github.com/tock/tock/pull/3802 > - Signature HIL - https://github.com/tock/tock/pull/3772 > - libtock-c refresh - How to make it more user friendly? > - What types of changes would make libtock-c easier to use? Where > are the pain points? > - Do we need yield-wait-for-x? > - Should we copy some existing interface? > - We probably aren't going to be able to do an entire rewrite, but > are there manageable steps that would make a difference? > > > On Fri, Feb 2, 2024 at 11:24 AM Brad Campbell <bradjc(a)virginia.edu> wrote: > >> >> - Process Checking Infrastructure PRs >> - https://github.com/tock/tock/pull/3772 >> - https://github.com/tock/tock/pull/3818 >> - Remove a nightly feature >> - https://github.com/tock/tock/pull/3803 >> - Finalize Legal TRD >> - https://github.com/tock/tock/pull/3821 >> >> >> On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: >> >>> Please reply with agenda items and +1s for today's meeting. >>> >>> Hudson is away today, so I'll be leading, hence the later than usual >>> request for agenda items. Luckily I think there are plenty. I suggest at >>> least: >>> >>> - Brad's suggestion for a new documentation working group (there is >>> already approval for this, but we can maybe discuss whether/how to support >>> this kind of process more generally. could be very brief) >>> _______________________________________________ >>> Devel mailing list -- devel(a)lists.tockos.org >>> To unsubscribe send an email to devel-leave(a)lists.tockos.org >>> >>

Feb. 16, 2024

Re: Call for Tock meeting agenda items

by Brad Campbell

Ideas for today: - Stable Rust - https://github.com/tock/tock/pull/3802 - Signature HIL - https://github.com/tock/tock/pull/3772 - libtock-c refresh - How to make it more user friendly? - What types of changes would make libtock-c easier to use? Where are the pain points? - Do we need yield-wait-for-x? - Should we copy some existing interface? - We probably aren't going to be able to do an entire rewrite, but are there manageable steps that would make a difference? On Fri, Feb 2, 2024 at 11:24 AM Brad Campbell <bradjc(a)virginia.edu> wrote: > > - Process Checking Infrastructure PRs > - https://github.com/tock/tock/pull/3772 > - https://github.com/tock/tock/pull/3818 > - Remove a nightly feature > - https://github.com/tock/tock/pull/3803 > - Finalize Legal TRD > - https://github.com/tock/tock/pull/3821 > > > On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: > >> Please reply with agenda items and +1s for today's meeting. >> >> Hudson is away today, so I'll be leading, hence the later than usual >> request for agenda items. Luckily I think there are plenty. I suggest at >> least: >> >> - Brad's suggestion for a new documentation working group (there is >> already approval for this, but we can maybe discuss whether/how to support >> this kind of process more generally. could be very brief) >> _______________________________________________ >> Devel mailing list -- devel(a)lists.tockos.org >> To unsubscribe send an email to devel-leave(a)lists.tockos.org >> >

Feb. 9, 2024

Tock meeting agenda 02/02

by Amit Levy

- Updates - Process Checking Infrastructure PRs: - https://github.com/tock/tock/pull/3772 - https://github.com/tock/tock/pull/3818 - Remove a nightly feature: https://github.com/tock/tock/pull/3803 - Finalize Legal TRD: https://github.com/tock/tock/pull/3821 - How to create new WGs and how to change membership for a WG? Link: https://meetings.dialpad.com/room/tock/1vnk9r3e1b

Feb. 2, 2024

Re: Call for Tock meeting agenda items

by Brad Campbell

- Process Checking Infrastructure PRs - https://github.com/tock/tock/pull/3772 - https://github.com/tock/tock/pull/3818 - Remove a nightly feature - https://github.com/tock/tock/pull/3803 - Finalize Legal TRD - https://github.com/tock/tock/pull/3821 On Fri, Feb 2, 2024 at 11:07 AM Amit Levy <amit(a)amitlevy.com> wrote: > Please reply with agenda items and +1s for today's meeting. > > Hudson is away today, so I'll be leading, hence the later than usual > request for agenda items. Luckily I think there are plenty. I suggest at > least: > > - Brad's suggestion for a new documentation working group (there is > already approval for this, but we can maybe discuss whether/how to support > this kind of process more generally. could be very brief) > _______________________________________________ > Devel mailing list -- devel(a)lists.tockos.org > To unsubscribe send an email to devel-leave(a)lists.tockos.org >

Feb. 2, 2024

Call for Tock meeting agenda items

by Amit Levy

Please reply with agenda items and +1s for today's meeting. Hudson is away today, so I'll be leading, hence the later than usual request for agenda items. Luckily I think there are plenty. I suggest at least: - Brad's suggestion for a new documentation working group (there is already approval for this, but we can maybe discuss whether/how to support this kind of process more generally. could be very brief)

Feb. 2, 2024