Re: [Emc-developers] loadusr -W

2025-07-18 Thread andy pugh
On Fri, 18 Jul 2025 at 23:59, rodw wrote: > > It does not make sense to push changes back to 2.9 unless you intend to > release a 2.9.5 I can't predict the future, but the policy is that bug fixes go into 2,9, and this is a bug fix. 2.10 is likely to include some large changes, and so isn't fee

Re: [Emc-developers] loadusr -W

2025-07-18 Thread rodw
It does not make sense to push changes back to 2.9 unless you intend to release a 2.9.5 But that makes no sense given the existing 2.9.4 release is in Debian trixie which is frozen pending a full Debian release in the next 60 days or so. Wouldn't it be best to stay in step with Debian with our ve

Re: [Emc-developers] loadusr -W

2025-07-18 Thread andy pugh
On Fri, 18 Jul 2025 at 23:02, Bertho Stultiens wrote: > An additional test can be in runtests(.in): > > wait_for_result_close() { > Would you like to make that a PR into 2.9? I think that that will make CI run it in the feature branch, so we can see how it goes on the serversm and I ca

Re: [Emc-developers] loadusr -W

2025-07-18 Thread Bertho Stultiens
On 7/18/25 10:13 PM, andy pugh wrote: The same race condition is still present because the 'result' file is closed (in runtests by the shell) when halrun exists. But we know that halrun's exit is not enough to prevent the race because halrun must have exited or 'checkresult' will not be executed.

Re: [Emc-developers] loadusr -W

2025-07-18 Thread andy pugh
On Fri, 18 Jul 2025 at 16:55, Bertho Stultiens wrote: > another race condition. > > In mb2hal.c: > hal_ready(gbl.hal_mod_id); > OK(gbl.init_dbg, "HAL components created OK"); > gbl.quit_flag = 0; > signal(SIGINT, quit_signal); > signal(SIGTERM, quit_signal); Yes, that do

Re: [Emc-developers] loadusr -W

2025-07-18 Thread Bertho Stultiens
On 7/18/25 2:19 PM, andy pugh wrote: That is interesting. So, why/how does mb2hal exit... I think that halrun -f always exits once the HAL file is parsed unless the -I (interactive) flag is set: https://linuxcnc.org/docs/stable/html/man/man1/halrun.1.html "If a filename is specified and neither

Re: [Emc-developers] loadusr -W

2025-07-18 Thread andy pugh
On Fri, 18 Jul 2025 at 10:16, Bertho Stultiens wrote: > That is interesting. So, why/how does mb2hal exit... I think that halrun -f always exits once the HAL file is parsed unless the -I (interactive) flag is set: https://linuxcnc.org/docs/stable/html/man/man1/halrun.1.html "If a filename is spe

Re: [Emc-developers] loadusr -W

2025-07-18 Thread Bertho Stultiens
On 7/17/25 10:10 PM, andy pugh wrote: You need to change the '-W' into '-w' (lower case 'w'). That should I will try that, though I am currently unclear what sends the EXIT I tried this, and it doesn't work as the HAL file just sits there waiting for the component to exit, but there is no exit