Fixes and improvements for Linux by BjoernAtBosch · Pull Request #25 · eclipse-score/time

BjoernAtBosch · 2026-05-21T14:35:45Z

This PR fixes some issues discovered in Linux (closes #36):

Fix reception of PTP frames: Previous filtering mechanism in raw_socket.cpp wasn't working under Linux
Add (temporary) cmdline parsing to allow configuration of Ethernet interface to be used
Use kernel sw timestamps as fallback if hw timestamps are not supported
Extend logging

github-actions · 2026-05-21T14:36:53Z

License Check Results

🚀 The license check job ran with the Bazel command:

bazel run //:license-check

Status: ⚠️ Needs Review

Click to expand output

[License Check Output]
Extracting Bazel installation...
Starting local Bazel server (8.6.0) and connecting to it...
INFO: Invocation ID: 68c96304-5f9f-4d3b-8420-8d2353089aea
Computing main repo mapping: 
WARNING: /home/runner/work/time/time/MODULE.bazel:13:7: The attribute 'compatibility_level' in module() is a no-op and will be removed in a future Bazel release. Please remove it from your MODULE.bazel file.
Computing main repo mapping: 
Computing main repo mapping: 
Computing main repo mapping: 
Computing main repo mapping: 
Loading: 
Loading: 3 packages loaded
Loading: 3 packages loaded
    currently loading: 
Loading: 3 packages loaded
    currently loading: 
Loading: 3 packages loaded
    currently loading: 
Analyzing: target //:license-check (4 packages loaded, 0 targets configured)
Analyzing: target //:license-check (4 packages loaded, 0 targets configured)

Analyzing: target //:license-check (61 packages loaded, 10 targets configured)

Analyzing: target //:license-check (141 packages loaded, 1065 targets configured)

Analyzing: target //:license-check (153 packages loaded, 6128 targets configured)

Analyzing: target //:license-check (153 packages loaded, 9185 targets configured)

Analyzing: target //:license-check (154 packages loaded, 9205 targets configured)

Analyzing: target //:license-check (165 packages loaded, 9284 targets configured)

Analyzing: target //:license-check (165 packages loaded, 9284 targets configured)

Analyzing: target //:license-check (166 packages loaded, 9284 targets configured)

Analyzing: target //:license-check (168 packages loaded, 11172 targets configured)

INFO: Analyzed target //:license-check (170 packages loaded, 11422 targets configured).
[11 / 14] checking cached actions
[12 / 16] JavaToolchainCompileClasses external/rules_java+/toolchains/platformclasspath_classes; 0s disk-cache, processwrapper-sandbox ... (2 actions running)
INFO: From Generating Dash formatted dependency file ...:
INFO: Successfully converted 2 packages from Cargo.lock to bazel-out/k8-fastbuild/bin/formatted.txt
[14 / 16] JavaToolchainCompileBootClasspath external/rules_java+/toolchains/platformclasspath.jar; 0s disk-cache, processwrapper-sandbox
INFO: Found 1 target...
Target //:license.check.license_check up-to-date:
  bazel-bin/license.check.license_check
  bazel-bin/license.check.license_check.jar
INFO: Elapsed time: 38.402s, Critical Path: 2.29s
INFO: 16 processes: 12 internal, 3 processwrapper-sandbox, 1 worker.
INFO: Build completed successfully, 16 total actions
INFO: Running command line: bazel-bin/license.check.license_check ./formatted.txt <args omitted>
usage: org.eclipse.dash.licenses.cli.Main [-batch <int>] [-cd <url>]
       [-confidence <int>] [-ef <url>] [-excludeSources <sources>] [-help] [-lic
       <url>] [-project <shortname>] [-repo <url>] [-review] [-summary <file>]
       [-timeout <seconds>] [-token <token>]

github-actions · 2026-05-21T14:41:16Z

The created documentation from the pull request is available at: docu-html

gordon9901 · 2026-07-09T10:25:21Z

    Close();

-    const int fd = sys_->socket_call(AF_PACKET, SOCK_RAW, htons(ETH_P_1588));
+    const int fd = sys_->socket_call(AF_PACKET, SOCK_RAW, htons(ETH_P_ALL));


The BPF filter only checks EtherType at offset 12 == 0x88F7, but VLAN-tagged frames have 0x8100 there — the inner 0x88F7 is at offset 16. Those frames are silently dropped before reaching frame_codec, which already handles them correctly. If VLAN-tagged PTP traffic is not expected in the target environment this is fine,

With the previous configuration, I wasn't able to receive gPTP frames in the time_slave. That's the reason to change it. Any gPTP frame should not be VLAN tagged - see my reply to Valery's comment.

gordon9901 · 2026-07-09T10:39:12Z

+
+    for (++iter; iter != args.cend(); ++iter)
+    {
+        if (*iter == "-h" || *iter == "--help")


Returning kInitFailure for --help is problematic — in an automotive lifecycle framework this can trigger watchdog restarts or fault reports. --help is intentional, not an error. Consider exit(0) directly after printing usage, or a separate return code.

I can change that, but not sure if returning a 0 (zero) is the right thing. Documentation of the Initialize function tells:

If non-zero returned, application will exit with exit
code returned from Initialize and Run method won't be called.

That's not an issue. That is what we want in case help was requested. But:

NOTE: In case non-zero is returned, the whole group where this
application belongs to will not leave the machine state Init and
thus, all the applications of such group will never be promoted to
the machine state Running. This could lead to the whole ECU not being
able to startup at all which will result in continuous general rejects.

Could it cause issues if we return a zero here from the terminated process (signalling success)? Normally, I'd expect that it should not because the process is expected to be running ...

I mean, this is a bit artificial because in a correctly configured system the time_slave shouldn't be called with -h or --help.

gordon9901 · 2026-07-09T10:40:34Z

 {

-constexpr int kRxTimeoutMs = 100;  // poll timeout; keeps RxLoop responsive to shutdown
+constexpr int kRxTimeoutMs = 500;  // poll timeout; keeps RxLoop responsive to shutdown


kRxTimeoutMs raised from 100 → 500 ms with no explanation in the commit message. Worst-case shutdown latency for RxLoop is now 5× longer. Please document the reason (CPU load? NIC behaviour?) so the timing impact can be assessed.

My intention was to reduce checking load and 0.5 sec isn't a long period. But we can switch back to the previous value if preferred.

Understood. Worth noting though that gPTP Sync messages arrive every ~125 ms, so a 500 ms poll wait could span 4 Sync cycles. Not a bug, just something to keep in mind.

lavrovvalera · 2026-07-09T09:35:26Z

+              << " [options]\n"
+                 "Options:\n"
+                 "  -h, --help Print this message and exit.\n"
+                 "  -i, --interface <iface> Define the ethernet interface to be used (default is \""


Do we have more options, which are needed for the slave? I would assume, yes
shall we specify then the config file and make it mandatory to be run?

My intention was to solve the issue first that the interface name differs between QNX and Linux. Furthermore it is quite likely that the user wants to use another interface.
Further options could be added as needed.

lavrovvalera · 2026-07-09T09:36:59Z

+    auto iter = args.cbegin();
+    const auto& app_name = *iter;
+
+    for (++iter; iter != args.cend(); ++iter)


Do we have already and command parses solutions in the base libs already?

Seems not: Neither lifecycle mgm has nor I found some in baselibs.

lavrovvalera · 2026-07-09T10:40:29Z

+    // (outer EtherType is 0x8100, not 0x88F7). The BPF filter runs in-kernel before delivery
+    // to userspace, so non-PTP frames are dropped with zero overhead in the application.
+    static const ::sock_filter kPtpBpfCode[] = {
+        BPF_STMT(BPF_LD | BPF_H | BPF_ABS, 12),                    // load EtherType


Can we change the "12" to some not -magic number?

lavrovvalera · 2026-07-09T10:41:04Z

+    static const ::sock_filter kPtpBpfCode[] = {
+        BPF_STMT(BPF_LD | BPF_H | BPF_ABS, 12),                    // load EtherType
+        BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, ETH_P_1588, 1, 0),     // == 0x88F7 → PASS
+        BPF_STMT(BPF_RET | BPF_K, 0U),                             // FAIL: drop


the same here and on the line below

Do you mean the 1 and the 0 in line 122? That's just the jump width to the filter instruction to be executed next ...

lavrovvalera · 2026-07-09T10:42:13Z

 {

-constexpr int kRxTimeoutMs = 100;  // poll timeout; keeps RxLoop responsive to shutdown
+constexpr int kRxTimeoutMs = 500;  // poll timeout; keeps RxLoop responsive to shutdown


why was it encreased? any justifications?

See reply to gordon9901's comment.

lavrovvalera · 2026-07-09T10:42:46Z

+        if (*iter == "-h" || *iter == "--help")
+        {
+            PrintUsage(app_name);
+            return false;


I would believe, it shall return true here

But would it terminate then? Typical behavior if querying help of a cmd is to print usage and terminate.

lavrovvalera · 2026-07-09T10:44:54Z

+{
+    const auto& args = context.get_arguments();
+    auto iter = args.cbegin();
+    const auto& app_name = *iter;


shall we check if args. are empty before resolving the iterator?
ApplicationContxt might not be == to the classical C argv*

Current impl is. It stores all C args in the vector. It has a separate private variable m_app_path to store the app path, but no public getter :-O

lavrovvalera · 2026-07-09T10:48:59Z


 void RawSocketImpl::Close()
 {
    const int fd = fd_.exchange(-1, std::memory_order_acq_rel);


Shall we drop here the PACKET_DROP_MEMBERSHIP for symmetry?

lavrovvalera · 2026-07-09T11:04:20Z

+    }
+
+    // BPF filter: pass only frames with EtherType 0x88F7 (PTP/gPTP).
+    // ETH_P_ALL is used for socket/bind because ETH_P_1588 misses VLAN-tagged PTP frames


the filter above, if I'm correct, handles the non tagged frames only.
to address vlan youwould need additioanl instructions.

but in general, I would believe the ptp frames shall not be vlan tagged, so the fileter is correct.
jsut the comments shall be adjusted

but in general, I would believe the ptp frames shall not be vlan tagged, so the fileter is correct.

Yes, gPTP specification (IEEE Std 802.1AS-2011) states in section 11.3.3: "A frame that carries an IEEE 802.1AS message shall not have a VLAN tag nor a priority tag."

Will adjust comment ...

ryan-steel · 2026-07-09T12:41:09Z

+    // ETH_P_ALL is used for socket/bind because ETH_P_1588 misses VLAN-tagged PTP frames
+    // (outer EtherType is 0x8100, not 0x88F7). The BPF filter runs in-kernel before delivery
+    // to userspace, so non-PTP frames are dropped with zero overhead in the application.
+    static const ::sock_filter kPtpBpfCode[] = {


kPtpBpfCode value (not type, qnx one is obviously qnx-specific) is identical to kPtp1588FilterInsns defined in gptp/platform/qnx/qnx_raw_shim.cpp - can we combine these somewhere common?

Don't know if that makes sense as this raw_socket impl is Linux-specific

The socket is, but the EtherType for PTP/gPTP is the same across platforms no?

Ah - got it now. But no, they are not exactly the same. Furthermore, these setting are quite driver-specific, so I'd like to keep it as is for now.

Ok - both filter instructions do the same. (Wasn't aware how the filter is exactly working - that part was fixed by AI). But still - I think this is quite OS specific, so I'd keep those separatly

github-project-automation Bot added this to TIM - Time FT May 21, 2026

github-project-automation Bot moved this to In Progress in TIM - Time FT May 21, 2026

BjoernAtBosch had a problem deploying to workflow-approval May 21, 2026 14:35 — with GitHub Actions Failure

BjoernAtBosch force-pushed the bjoernatbosch_get_it_running branch from 06f0df5 to b2de800 Compare June 15, 2026 10:41

BjoernAtBosch requested a deployment to workflow-approval June 15, 2026 10:41 — with GitHub Actions Waiting

BjoernAtBosch mentioned this pull request Jun 15, 2026

Replace mmap-based IPC with score_baselibs SharedMemoryFactory in ts_client #54

Merged

BjoernAtBosch added 7 commits July 8, 2026 10:52

Fix TimeSlave for Linux and extend logging

ec0754e

Use gptp_shm_machine instead of stub

a4c8078

Log incoming PTP messages

f5ba6f5

Use sw kernel timestamps as fallback to hw timestamps

5757b25

Fix rebase misses

23a48f7

Some fixes

0f9fe8a

Add cmdline parser for ethernet interface

a93e5aa

BjoernAtBosch force-pushed the bjoernatbosch_get_it_running branch from b2de800 to a93e5aa Compare July 8, 2026 12:30

BjoernAtBosch had a problem deploying to workflow-approval July 8, 2026 12:30 — with GitHub Actions Error

Clang-format

50f3ce9

BjoernAtBosch had a problem deploying to workflow-approval July 8, 2026 13:25 — with GitHub Actions Error

BjoernAtBosch had a problem deploying to workflow-approval July 8, 2026 13:44 — with GitHub Actions Error

Improve method name

949c02a

BjoernAtBosch force-pushed the bjoernatbosch_get_it_running branch from f201317 to 949c02a Compare July 8, 2026 13:58

BjoernAtBosch deployed to workflow-approval July 8, 2026 13:58 — with GitHub Actions Active

BjoernAtBosch temporarily deployed to workflow-approval July 8, 2026 13:58 — with GitHub Actions Inactive

BjoernAtBosch changed the title ~~Try to get it running in Linux~~ Fixes and improvements for Linux Jul 8, 2026

BjoernAtBosch marked this pull request as ready for review July 8, 2026 14:05

BjoernAtBosch requested review from LudwigWeise, gordon9901, lavrovvalera and ryan-steel and removed request for gordon9901 July 8, 2026 14:05

gordon9901 reviewed Jul 9, 2026

View reviewed changes

lavrovvalera reviewed Jul 9, 2026

View reviewed changes

ryan-steel reviewed Jul 9, 2026

View reviewed changes

gordon9901 approved these changes Jul 10, 2026

View reviewed changes

Fix BPF comment

b9c5a74

BjoernAtBosch requested a deployment to workflow-approval July 10, 2026 13:19 — with GitHub Actions Waiting

Uh oh!

Conversation

BjoernAtBosch commented May 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented May 21, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

License Check Results

Uh oh!

github-actions Bot commented May 21, 2026

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BjoernAtBosch Jul 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BjoernAtBosch Jul 9, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BjoernAtBosch Jul 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

BjoernAtBosch Jul 10, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Reviewers

Assignees

Labels

BjoernAtBosch commented May 21, 2026 •

edited

Loading

github-actions Bot commented May 21, 2026 •

edited

Loading

BjoernAtBosch Jul 10, 2026 •

edited

Loading

BjoernAtBosch Jul 9, 2026 •

edited

Loading

BjoernAtBosch Jul 10, 2026 •

edited

Loading

BjoernAtBosch Jul 10, 2026 •

edited

Loading