CCSDS AOS Framer by Willmac16 · Pull Request #4630 · nasa/fprime

Willmac16 · 2026-01-17T20:26:45Z


*Related Issue(s)*
*Has Unit Tests (y/n)*	y
*Documentation Included (y/n)*	y
*Generative AI was used in this contribution (y/n)*	n

Change Description

Implements a CCSDS AOS Framer that supports both many packets per frame and packets than span multiple frames.

Rationale

AOS is a newer protocol than TM (although USLP is even newer). Requiring that packets fit within one frame is overly restrictive.

Testing/Review Recommendations

Ensure code style, layout and coverage looks good. Determine if existing subset of AOS features would be sufficient for Fprime downlink.

Future Work

An AOS Deframer will be made to accompany this, and AOS gds infrastructure will also likely be added.

AI Usage (see policy)

N/A

Willmac16 · 2026-01-17T20:27:28Z

Only uncovered line is the (not yet implemented) EPP Idle packet generation

1/1 Test #62: Svc_Ccsds_AosFramer_ut_exe .......   Passed    0.09 sec

The following tests passed:
        Svc_Ccsds_AosFramer_ut_exe

100% tests passed, 0 tests failed out of 1

Total Test time (real) =   6.33 sec
(INFO) Reading coverage data...
(INFO) Writing coverage report...
lines: 99.5% (205 out of 206)
functions: 100.0% (23 out of 23)
branches: 49.0% (423 out of 864)

thomas-bc

I haven't been able to dive in the code itself too deep yet, as I was also discovering the AOS spec in parallel. I should be able to get to it more thoroughly in the next day or two. But in the meantime, for the sake of iterating more quickly, I figured I'd dump a first set of nits/questions/comments.

This looks great! Love the internal pattern of tracking AosVc's to extend further down the road.

cmake/platform/Darwin.cmake

.vscode/launch.json

Svc/Ccsds/Types/Types.fpp

Svc/Ccsds/AosFramer/AosFramer.cpp

default/config/ComCfg.fpp

Svc/Ccsds/AosFramer/AosFramer.cpp

Svc/Ccsds/AosFramer/docs/sdd.md

Svc/Ccsds/AosFramer/AosFramer.cpp

github-advanced-security

CodeQL found more than 20 potential problems in the proposed changes. Check the Files changed tab for more details.

Svc/Ccsds/AosFramer/AosFramer.cpp

thomas-bc · 2026-01-30T00:05:53Z

yeah those RHEL8 warnings are annoying, but we've seen that before, just need to cast the whole result of the operation... for example
https://github.com/willmac16/fprime/blob/760ca7c4554b2241906fac6ca2b7ba868ee13d9a/Svc/Ccsds/SpacePacketDeframer/test/ut/SpacePacketDeframerTester.cpp#L50-L51

Svc/Ccsds/AosFramer/AosFramer.cpp

thomas-bc

Code LGTM; just a few follow-ups and questions in some places.

default/config/ComCfg.fpp

Svc/Ccsds/AosFramer/AosFramer.cpp

default/config/ComCfg.fpp

thomas-bc · 2026-02-03T01:52:46Z

default/config/ComCfg.fpp

-        vcId: U8                    @< 6 bit Virtual Channel ID - used for TC and TM
+        vcId: U8                    @< 6 bit Virtual Channel ID - used for AOS
+        replayFlag: bool            @< AOS bool for realtime vs replay (passed directly into the frame)
+        sendNow: bool               @< Flag to AOS Framer that the Frame this packet goes into should be sent ASAP


Could this functionality come in on a separate port instead? What component do you envision would turn that on in the framing chain?

My thought process is that some sources upstream of the ComQueue might want to request immediate sending, while others would prioritize bandwidth over latency. Since we're already conveying the APID via the context it seems reasonable to me that we'd convey sending immediacy this way too

thomas-bc · 2026-02-03T02:00:47Z

Svc/Ccsds/AosFramer/docs/sdd.md

General question; are you all planning to use AOS for a specific reason that you would be willing to share?

I have been looking at these standards for a bit but am still out of the loop on use-cases and such. For example, why use AOS and not USLP? Is AOS better supported by other tools because of its "seniority" over USLP ?

From what I understand, AOS Downlink + TC Uplink is pretty common. Or you can go USLP (In different modes I believe) both ways.

For FPrime upstream I think it makes sense to swap the ComAggregator (which I didn't know about until today) + TM Framer for the AOS Framer since it enables packets that span multiple frames. Or at least make a new AOS+TC Subtopo that does. And if FPrime gets a USLP defamer that can do variable length deframing, then it would make sense to switch over/provide USLP for both.

The big AOS upgrade--that this implementation doesn't provide--is the transfer frame insert zone, which is meant for realtime voice/video to be interspersed at a regular rate w/ standard telemetry & downlink.

We are likely not using this AOS framer for anything other than payload comms due to particulars of our radio.

thomas-bc · 2026-02-03T02:02:07Z

Svc/Ccsds/AosFramer/AosFramer.cpp

Has this Framer been tested end-to-end against a tool that is known to deframe AOS frames correctly?

Not yet. I could try to work out a YAMCS test for downlink; however, my primary need for this component is for communicating w/ payloads, so I was planning on placing my integration testing focus on comms w/ those particular payloads

length

thomas-bc · 2026-02-03T16:56:34Z

cmake/platform/Darwin.cmake

   BASE_CONFIG
 )
 target_compile_definitions(PlatformDarwin INTERFACE -DTGT_OS_TYPE_DARWIN)
-target_link_options(PlatformDarwin INTERFACE -Wl,-no_warn_duplicate_libraries)


We should revert this deletion, maybe appeared because of the force push or something

Svc/Ccsds/AosFramer/AosFramer.cpp

+
+AosFramer ::AosFramer(const char* const compName) : AosFramerComponentBase(compName) {
+    // Default to FECF on, Max Sized if you don't override w/ another configure call
+    configure(ComCfg::AosMaxFrameFixedSize, true);


Svc/Ccsds/AosFramer/AosFramer.cpp

+    // Default to FECF on, Max Sized if you don't override w/ another configure call
+    configure(ComCfg::AosMaxFrameFixedSize, true);
+}
+


Svc/Ccsds/AosFramer/AosFramer.cpp

+}
+
+// ----------------------------------------------------------------------
+// Handler implementations for typed input ports


Svc/Ccsds/AosFramer/AosFramer.cpp

+    }
+
+    // Pack this packet
+    pack_pad_send(data, context);


Svc/Ccsds/AosFramer/AosFramer.cpp

+    // Ensure configure was called
+    FW_ASSERT(currentVc.vc_struct_index == ind);
+    FW_ASSERT(currentVc.virtualChannelId == context.get_vcId());
+    return currentVc;


Svc/Ccsds/AosFramer/AosFramer.cpp

+    Fw::SerializeStatus status = serializer.serializeFrom(header);
+    FW_ASSERT(status == Fw::FW_SERIALIZE_OK, status);
+
+    // Fill with idle pattern


Svc/Ccsds/AosFramer/AosFramer.hpp

+        OWNED,      //!< The buffer is currently owned by the AosFramer
+    };
+
+    struct PDU {


Svc/Ccsds/AosFramer/AosFramer.hpp

+        FwSizeType offset = 0;         //!< Offset into the above packet to write
+    };
+
+    struct AosVc {


Svc/Ccsds/AosFramer/AosFramer.hpp

+
+        // Because the AOS protocol use fixed width frames, and only one frame is in transit between ComQueue and
+        // ComInterface at a time, we can use a member fixed-size buffer to hold the frame data
+        struct FrameBuffer {


Svc/Ccsds/AosFramer/AosFramer.hpp

+        // SPP Idle packet backstop
+        // Technically we'd only use 6 of the 7 bytes at worst
+        // cuz the first one had to go into the prev frame
+        struct SppIdle {


Svc/Ccsds/AosFramer/AosFramer.cpp

+// Component construction and destruction
+// ----------------------------------------------------------------------
+
+AosFramer ::AosFramer(const char* const compName) : AosFramerComponentBase(compName) {


Willmac16 changed the title ~~Aos framing~~ CCSDS AOS Framer Jan 17, 2026

thomas-bc self-requested a review January 20, 2026 17:38

thomas-bc reviewed Jan 21, 2026

View reviewed changes

github-advanced-security bot found potential problems Jan 21, 2026

View reviewed changes

Svc/Ccsds/AosFramer/AosFramer.cpp Fixed Show fixed Hide fixed

github-advanced-security bot found potential problems Jan 21, 2026

View reviewed changes

github-advanced-security bot found potential problems Jan 30, 2026

View reviewed changes

Svc/Ccsds/AosFramer/AosFramer.cpp Fixed Show fixed Hide fixed

github-advanced-security bot found potential problems Feb 2, 2026

View reviewed changes

Svc/Ccsds/AosFramer/AosFramer.cpp Fixed Show fixed Hide fixed

thomas-bc reviewed Feb 3, 2026

View reviewed changes

Willmac16 added 20 commits February 3, 2026 09:25

Init AOS (De)Framer from TC & TM

5ff3623

Add AOS Suppport Subset list to sdd

cb9e6a4

Start Refining AOS Framer User Interface

7ceb8f4

Move PVN to Types.fpp & Fill out TFVN

7f4fb67

AOS Framer Header

8a93680

Refactor the Virtual Channel as a struct

1acb389

More work on idle packet serialization

7cadcfa

Move PVN back into ComCfg

48f4f93

AosFramer Builds

a3ebe4e

More Section Number & Refactor Sending, Packing, & Header Writing logic

ec86e58

Start Cleaning up AosFramer w/ Unit Tests

d91a80e

Ensure SPP Idle Fill increments payload offset

060384e

Mask & shift (don't shift masks)

1aea426

Nothing too large & sendNow for buffer ownership test

43a4d5d

Fix SeqCount wrap around test

925acd1

Created, but failing long packet test

3c41505

Add 2 TODOs

88fba1c

Make AosFrmer Happy on aarch64

3510bd1

Add lldb vscode launch.json config

3a0bf8d

Passing striped packet test (Packet across multiple frames)

7421a9f

Willmac16 added 20 commits February 3, 2026 09:25

Apply a subset of the spelling bot patch

7da39da

fix re-entrancy and fight w/ aos spelling pattern

255a041

just aos

6f5fcf0

aos -> Aos

e615192

Revert launch.json and macos library duplicate change in other PR

20e3019

Correct annotations for PVN dict

40ccc25

Make Spacing uniform for ComCfg.fpp annotations

ee37e6d

Rename compute_fecf to compute_and_inject_fecf

f46b0dd

Correct sdd's description of provided service to VCP

04b6fd6

Add a pair of asserts to try to make RHEL compile CICD happy

98a89e2

Add pack_pad_send helper to avoid pack_packet reentrancy

cd9eb5b

Update sdd to point at version 5 of the AOS SDL Standard

39b5755

Default FECF on & constant driven AOS Frame Size

da57e04

More RHEL8 appeasal

8734acc

RHEL8 UT Appeasement

2f47c9a

VC ID default to 1, replay flag gone, & serialize idle spp takes a U16

bab9756

length

Improve comment about 24 + 4 bit frame count

71a2aa3

More RHEL8 UT complaints

cbab740

expect.txt bitfield

40b108a

Back to expecting Bitfield

fe4254c

Willmac16 force-pushed the aos-framing branch from 2772cbe to fe4254c Compare February 3, 2026 15:26

Willmac16 added 3 commits February 3, 2026 09:33

Shift then Mask to hopefully silence RHEL UT

2492f42

spelling

b6da646

Try Unsigned masks

59c7bb3

thomas-bc reviewed Feb 4, 2026

View reviewed changes

Willmac16 added 2 commits February 4, 2026 08:23

Add back in Darwin linker dupe warning suppression

bd39f50

Promote to unsigned int before masking or shifting

25c02a8

github-advanced-security bot found potential problems Feb 4, 2026

View reviewed changes

Hope that U32 cast is enough to ward of promotion

739ebf0

Conversation

Willmac16 commented Jan 17, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Change Description

Rationale

Testing/Review Recommendations

Future Work

AI Usage (see policy)

Uh oh!

Willmac16 commented Jan 17, 2026

Uh oh!

thomas-bc left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

github-advanced-security bot left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

thomas-bc commented Jan 30, 2026

Uh oh!

Uh oh!

thomas-bc left a comment

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

thomas-bc Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

Willmac16 Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

thomas-bc Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

Willmac16 Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

thomas-bc Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

Willmac16 Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

thomas-bc Feb 3, 2026

Choose a reason for hiding this comment

Uh oh!

Check notice

Check notice

Check notice

Check notice

Check notice

Check notice

Check notice

Check notice

Check notice

Check notice

Check notice

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

2 participants

Willmac16 commented Jan 17, 2026 •

edited

Loading